test(validate-skills): document test helpers with JSDoc

Added JSDoc to the test/assert/hasTriggerFinding helpers so the new
test file documents its own helpers. No behaviour change — the suite
still passes 3/3.

.gitignore

@@ -47,8 +47,6 @@ CLAUDE.local.md
 .claude/settings.local.json
 .junie/
 .agents/
-.analysis/
-
 
 z*/
 !docs/zh-cn/

CHANGELOG.md

@@ -1,46 +1,5 @@
 # Changelog
 
-## v6.9.0 - 2026-06-21
-
-### ✨ Headline
-
-**Reasoning skills get sharper and orchestration gets a memory.**
-
-**bmad-forge-idea** is a new core skill that takes a half-formed idea and pressure-tests it one Socratic question at a time — with an adversarial attack mode and optional persona rooms — until the idea hardens, proves out, or dies cheaply. 
-
-**bmad-architecture** lands as a ground-up rewrite of the old multi-step create-architecture flow: a lean spine (`ARCHITECTURE-SPINE.md`) that is the source of truth, intent-based routing (Create/Update/Validate), a breadth-coverage rubric so no dimension is silently skipped, and an opt-in reviewer gate.
-
-**party-mode** is reborn with creatable, savable custom parties, optional party memory, and many pacing and dynamics improvements.
-
-**Under the hood:** a canonical shared **memlog** (`_bmad/scripts/memlog.py`) replaces per-skill decision logs and is now the standard working-memory primitive across the suite. The installer now checks for **uv** and reframes it as the standard way to run BMAD's Python scripts (`uv run`). Plus an **Astro 6** security upgrade clearing 8+ Dependabot advisories and two new platform targets.
-
-### ⚠️ Upcoming Breaking Change (in v7) — standardizing on `uv`
-
-The industry is converging on [**uv**](https://docs.astral.sh/uv/) for running Python, and BMAD is following. Today our skills use a **mix** of `uv run` and direct `python3` invocation. In the **v7 release, every skill that runs a Python script will standardize on `uv run`** instead of calling `python3` directly — `uv` provisions the interpreter and manages dependencies, so scripts run consistently regardless of what's on your PATH.
-
-**What to do now:** install and set up `uv` ([docs](https://docs.astral.sh/uv/)) — or just ask your AI agent to "install and set up uv for me." Starting this release the installer checks for it and points you to setup if it's missing. `uv` is **not yet required** but without it some skills may have degraded performance or a shim AGENTS.md (or similar) or rule will need to be added to your environment to tell the agent when it sees uv run to use python3 instead. The best course of action though at this time is to install uv. A missing `uv` still warns rather than blocks, but it will be the assumed default in v7. Custom skills and overrides that shell out to `python3` should plan to migrate to `uv run`.
-
-### 🎁 Features
-
-* **bmad-forge-idea — new core skill** (#2492). Domain-agnostic idea pressure-testing for the analysis phase: Socratic, one-question-at-a-time interrogation with an adversarial attack mode and optional persona rooms resolved from the installed roster. Hardens or kills an idea cheaply; emits memlog residue and an optional brief that feeds bmad-spec or bmad-quick-dev. Interactive only (menu code FI).
-* **bmad-architecture — lean spine rewrite** (#2467, #2475). Replaces the fixed-step `bmad-create-architecture` (retained as a forwarding shim, removed in v7) with intent-based routing across five entry shapes (raw idea, large doc, codebase, feature slice, existing spine). The spine (`ARCHITECTURE-SPINE.md`) is the source of truth and SPEC.md is derived from it. Adds a breadth-coverage rubric (every altitude-owned dimension decided/deferred/open), an opt-in reviewer gate that scales lenses to rigor, and a full non-interactive headless mode. `lint_spine.py` hardened with fence-blanking, robust column detection, and 28 regression tests.
-* **party-mode: configurable parties + persistent memory** (#2479, #2484). Custom personas (`party_members`) and named rooms (`party_groups`, with optional scenes), four run modes (auto/session/subagent/agent-team), and a preloaded "Code Review Crew" of five adversarial lenses. Each party keeps append-only session memory under `{memory_dir}/<party_id>/` so sessions resume with prior context; ad-hoc casts stay ephemeral.
-* **bmad-brainstorming: facilitation modes + visual composer** (#2445). Three modes (Facilitator / Creative Partner / Ideate for me), append-only memlog with optional `--by` authorship attribution, and a self-contained `brain-selector.html` composer (technique strategy, category chips, filter, copy-to-clipboard, dark mode). Catalog grows to 108 techniques (8 new classics: HMW, JTBD, Empathy Map, Backcasting, TRIZ, Fishbone, Build on What Works, Scenario Cross) plus a convergence phase.
-* **Canonical shared memlog script** (#2462). New `src/scripts/memlog.py` — append-only chronological working memory with init/append/set ops, no lifecycle-status design, Python 3.8+ support, 30 tests. Any skill can call it at runtime.
-* **Retrospective action items tracked in sprint-status** (#2465). The retrospective step appends an `action_items` section to `sprint-status.yaml`; sprint-status validates and surfaces open items, and sprint-planning preserves them on regenerate.
-* **Installer checks for `uv` and reframes it as the standard** (#2495). Replaces the old python3 probe with a `uv` check, adds a heads-up to the install intro and a tip to the "BMAD is ready" summary, and updates docs/script docstrings (en/fr/vi-vn) to frame `uv run` as the standard and `python3` as the transition fallback. Migration-friendly: a missing `uv` warns and points you to setup, never blocks. See the Upcoming Breaking Change note above.
-* **New installer platform targets: hermes-agent and CodeWhale** (#2489, #2459). hermes-agent added as a tool target; CodeWhale uses `.codewhale/skills/` (project) and `~/.codewhale/skills/` (global), both with test coverage.
-
-### 🐛 Fixes
-
-* **Astro 6 security upgrade clears Dependabot alerts** (#2493). Astro 5.18.1 → 6.4.6 and Starlight 0.37.5 → 0.40.0 (8 XSS/SSRF advisories), esbuild pinned to 0.28.1 (Windows dev-server file read), markdown-it 14.2.0 (smartquotes ReDoS), brace-expansion 5.0.6 (range DoS). Docs content config migrated to `src/content.config.ts`; page output verified identical to baseline.
-* **Guard WSL installs from Windows Node** (#2470). Detects and prevents a Windows `node.exe` being used inside WSL, where it would silently fail.
-* **Remove empty skill-group dirs after install** (#2461). Prunes empty parent dirs (e.g. `_bmad/bmm/1-analysis`) left after skill cleanup, with a path-boundary check to avoid sibling-dir collisions.
-* **bmad-create-epics-and-stories discovers bmad-ux spine outputs** (#2446). Prerequisites now recognize `DESIGN.md` / `EXPERIENCE.md` alongside the legacy `ux-spec.md`.
-* **Pass diff inline to the blind-hunter reviewer** (#2463). Diff output is passed inline in the subagent prompt rather than via a file the reviewer can't read, preventing context-starved hallucination.
-* **Website: nav height for dual announcement banners** (#2473). Fixes layout crowding when two banners show at once.
-* **Workflow clarity & numbering** — clarify quick-dev subagent use across code-review/create-story/quick-dev (#2450), renumber retrospective steps (#2448).
-
 ## v6.8.0 - 2026-05-25
 
 ### ✨ Headline

docs/explanation/forge-idea.md

@@ -1,76 +0,0 @@
----
-title: "Forge an Idea"
-description: Pressure-test an idea through persona-driven interrogation until it hardens, proves out, or dies cheaply
-sidebar:
-  order: 14
----
-
-Take a half-formed idea and pressure-test it now, in conversation, while changing your mind is still free.
-
-## What is Forge Idea?
-
-Run `bmad-forge-idea` and an exacting interrogator goes to work on your idea, one question at a time, until what survives is something you can act on with earned conviction. The skill is domain-agnostic. It runs on a software feature, a business model, a research hypothesis, or a life decision you keep circling.
-
-What you walk away with is sharper thinking. A distilled `forged-idea.md` is only ever one possible exit, and the session never herds you toward "shall we build it?"
-
-## Why Pressure-Test Early
-
-The enemy is the hole you can't see in your own idea. An unexamined assumption or an unresolved branch is a crack, and a crack you miss now resurfaces later — in the build, or the launch, when it costs far more to fix.
-
-A conversation is the cheapest place to catch it, because changing your mind here costs nothing. The forge spends that cheapness on purpose, going after the weak points while fixing them is still free.
-
-## How a Session Runs
-
-The interrogator works one question at a time, in dependency order, and puts its own recommended answer on the table each time. A position you can push against gets further than an open prompt. It finds discoverable answers itself instead of sending you to fetch them.
-
-When your idea lands inside an existing project, that project's material becomes the ground truth. The interrogator checks your claims against what already exists and names the contradictions. Your vocabulary gets the same treatment. When a term is fuzzy or carries two meanings, it forces a precise choice before the branch can resolve, because a branch built on an overloaded word resolves falsely.
-
-## The Room
-
-The forge is voiced. Once the topic is set, every branch arrives with two characters instead of one faceless assistant. One comes from your installed roster — an agent or persona you'll recognize, drawn from the same cast behind [Party Mode](./party-mode.md) and [named agents](./named-agents.md). The other is conjured on the fly by the topic itself: a hostile competitor, a skeptical CFO, a domain specialist who has watched this exact plan fail before.
-
-You steer the room whenever you want. Name a specific person, call a saved party, or invoke the **adversarial on this** gear to attack a claim to destruction with you defending it.
-
-## Never Default-Agree
-
-Reflexive agreement is the failure this skill exists to refuse. Acknowledging your idea isn't the same as endorsing it, and the forge won't praise anything before it has survived something. It attacks the weak point or builds on the strong one, and it credits only what genuinely earns the credit.
-
-This is the deliberate inverse of [Adversarial Review](./adversarial-review.md). There, the reviewer is told to find problems and you filter out the false positives. Here, the interrogator is told never to grant agreement for free, so the pressure stays high and you think harder under it. It optimizes for the best idea over a comfortable session.
-
-## How a Session Ends
-
-The session ends however the thinking lands, and every landing is a real outcome. The forge writes a self-contained report you can keep, stamped to match the result.
-
-| Outcome | What it means |
-| --- | --- |
-| **Hardened** | The idea survived. It distills into `forged-idea.md` (the locked decisions, plus what was killed and why), ready to feed `bmad-spec`, `bmad-prd`, or `bmad-prfaq` for a product concept. |
-| **Killed** | The idea didn't survive, and the report records the cause of death. Finding that out cheaply is a win. |
-| **Clearer** | You just think straighter now. No artifact needed, and the session stands on its own. |
-
-## When to Use It
-
-Reach for the forge when you already hold an idea and want it hardened or killed before you invest in it. Skip it when you're still generating options or when you need a group decision from your agents.
-
-| Skill | Use it when | Voice |
-| --- | --- | --- |
-| `bmad-forge-idea` | You have an idea and want it stress-tested or killed | An interrogator plus a two-person room |
-| `bmad-prfaq` | You've committed to a product and want it proven customer-first | A Working Backwards coach |
-| `bmad-brainstorming` | You have no idea yet and need to generate options | A facilitation coach |
-| `bmad-party-mode` | You want your agents to discuss or decide together | Your whole roster in one conversation |
-| `bmad-review-adversarial-general` | You have an artifact and need its flaws found | A reviewer who must find issues |
-
-## Example
-
-:::note[Example]
-**You:** "I want AI-generated weekly summaries in the app. Users get a digest email every Monday."
-
-**Forge:** "Before the feature, who asked for this? My recommended answer is nobody did, and you're reasoning from 'AI summaries sound valuable.' Push back if there's a real signal."
-
-**Skeptical CFO (conjured):** "And a Monday digest is a recurring inference cost across your whole active base, forever. What revenue line pays for it?"
-
-**You:** "Retention, I guess. People forget to come back."
-
-**Forge:** "Then the load-bearing claim is 'a Monday email brings lapsing users back,' not 'AI summaries are valuable.' Those are different ideas. Which one are we forging?"
-:::
-
-The first idea was a feature. Two questions in, the real idea is a retention bet you could test with a plain email and no model at all.

docs/explanation/party-mode.md

@@ -5,7 +5,7 @@ sidebar:
   order: 11
 ---
 
-Party mode puts your AI agents in one room and lets them talk, to each other and to you. This page explains what a party is, the four ways it can run, how to build your own cast of personas instead of using the installed agents, and how a party remembers you between sessions.
+Party mode puts your AI agents in one room and lets them talk, to each other and to you. This page explains what a party is, the four ways it can run, and how to build your own cast of personas instead of using the installed agents.
 
 ## What is Party Mode?
 
@@ -131,16 +131,6 @@ Whichever mode is running, the orchestrator presents the result as one conversat
 You aren't limited to a single group. Pull members from several parties into the same conversation, or name a cast on the spot, and let them mix. Picture the Golden Girls thrown into an architecture review with Martin Fowler and Linus Torvalds, sparring over a change request: you can imagine how that goes.
 :::
 
-## The room remembers
-
-Give a party a memory and it picks up where you left off. It keeps its own record of your past sessions — the dynamics that built up between members, the threads you left open, and where earlier conversations landed. Reopen it a week later and that history is intact: two members who came to blows last time still open a little frosty, and a sharp line from a past session can resurface as an organic callback.
-
-It's memory, not a transcript. The room carries the few things worth remembering, not a log of everything said, so the next conversation feels continuous without dragging the whole past into it. It happens on its own, in the background — nothing to save, and the room never breaks character to announce it.
-
-A character who turns up on the fly is remembered too — a walk-on from an open-cast scene, or someone you add mid-conversation. At the end of a session the room offers to keep the new arrivals, folding them into the party so they can come back next time.
-
-Memory is set per party. When you create or save a party you're asked whether it should remember; the default installed-agent room remembers unless you turn it off. Set or change any of this through `/bmad-customize bmad-party-mode`.
-
 ## A keepsake of the session
 
 When you wrap up, the orchestrator offers a keepsake: a single self-contained HTML document of the session to keep or share. It lays the conversation out by persona rather than dumping a raw transcript. Decline it and the party simply ends.

docs/fr/how-to/customize-bmad.md

@@ -22,7 +22,7 @@ Le skill `bmad-customize` est un assistant de rédaction guidée pour les **opti
 :::note[Prérequis]
 
 - BMad installé dans votre projet (voir [Comment installer BMad](./install-bmad.md))
-- Un moyen d’exécuter le script de résolution — BMad adopte `uv` comme standard (`uv run`, qui provisionne Python pour vous) ; un simple `python3` 3.11+ sur votre PATH fonctionne toujours pendant la transition. Le script n’utilise que `tomllib` de la bibliothèque standard, il n’y a donc rien à `pip install`.
+- Python 3.11+ sur votre PATH (pour le script de résolution — utilise `tomllib` de la bibliothèque standard, pas de `pip install`, pas de `uv`, pas de virtualenv)
 - Un éditeur de texte pour les fichiers TOML
 :::
 
@@ -201,15 +201,15 @@ persistent_facts = [
 
 ## Comment fonctionne la résolution
 
-À l’activation, le SKILL.md de l’agent exécute un script Python partagé qui effectue la fusion à trois couches et renvoie le bloc résolu en JSON. Le script utilise uniquement le module `tomllib` de la bibliothèque standard Python (aucune dépendance externe). BMad adopte `uv run` comme standard pour exécuter ces scripts (uv provisionne un Python adapté pour vous) ; un simple `python3` fonctionne toujours pendant la transition :
+À l’activation, le SKILL.md de l’agent exécute un script Python partagé qui effectue la fusion à trois couches et renvoie le bloc résolu en JSON. Le script utilise le module `tomllib` de la bibliothèque standard Python (aucune dépendance externe), donc `python3` suffit :
 
 ```bash
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill {skill-root} \
   --key agent
 ```
 
-**Prérequis** : Python 3.11+ (les versions antérieures n’incluent pas `tomllib`). Rien à `pip install`. L’exécution via `uv run` est le standard à venir — uv résout un interpréteur adapté pour vous. Si vous l’exécutez directement avec `python3` pendant la transition, vérifiez votre version avec `python3 --version` ; certaines plateformes (macOS sans Homebrew, Ubuntu 22.04) ont `python3` par défaut en 3.10 ou antérieur, vous devrez peut-être installer 3.11+ séparément.
+**Prérequis** : Python 3.11+ (les versions antérieures n’incluent pas `tomllib`). Pas de `pip install`, pas de `uv`, pas de virtualenv. Vérifiez avec `python3 --version`. Certaines plateformes (macOS sans Homebrew, Ubuntu 22.04) ont `python3` par défaut en 3.10 ou antérieur, vous devrez peut-être installer 3.11+ séparément.
 
 `--skill` pointe vers le répertoire installé du skill (où se trouve `customize.toml`). Le nom du skill est déduit du basename du répertoire, et le script cherche automatiquement `_bmad/custom/{skill-name}.toml` et `{skill-name}.user.toml`.
 
@@ -217,17 +217,17 @@ Exemples d’utilisation :
 
 ```bash
 # Résoudre le bloc agent complet
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill /chemin/absolu/vers/bmad-agent-pm \
   --key agent
 
 # Résoudre un seul champ
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill /chemin/absolu/vers/bmad-agent-pm \
   --key agent.icon
 
 # Dump complet
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill /chemin/absolu/vers/bmad-agent-pm
 ```
 

docs/fr/reference/core-tools.md

@@ -113,7 +113,7 @@ La magie se produit dans les idées 50–100. Le workflow encourage la générat
 1. Lit l’entrée et tout document annexe lié
 2. Distille en un noyau à cinq champs via un modèle configurable ; redirige l’excédent vers des fichiers compagnons correctement nommés
 3. Exécute une auto-validation en deux passes (règles de cohérence, puis préservation de chaque affirmation essentielle de la source)
-4. Écrit `SPEC.md`, les compagnons associés, et un `.memlog.md` sous `{output_folder}/specs/spec-{slug}/`
+4. Écrit `SPEC.md`, les compagnons associés, et un `.decision-log.md` sous `{output_folder}/specs/spec-{slug}/`
 
 La loi Spec impose huit règles : les capacités expriment à la fois l’intention et le critère de succès ; les intentions décrivent le QUOI, pas le COMMENT ; les contraintes guident réellement les décisions ; les non-objectifs sont explicites ; les signaux de succès sont concrets ; les identifiants de capacité sont stables ; chaque affirmation essentielle de la source est préservée ; la rédaction est concise.
 
@@ -123,7 +123,7 @@ La loi Spec impose huit règles : les capacités expriment à la fois l’inten
 - `slug` (optionnel) — Requis uniquement lorsque l’entrée est succincte et qu’aucun slug ne peut être dérivé du nom de fichier source
 - `target_spec_path` (optionnel) — Définir pour mettre à jour une spécification existante au lieu d’en créer une nouvelle
 
-**Sortie :** Dossier de spécification contenant `SPEC.md`, les éventuels fichiers compagnons, et un `.memlog.md`. Les appelants en mode headless reçoivent une réponse JSON avec le statut du résultat et la liste des fichiers écrits ou modifiés.
+**Sortie :** Dossier de spécification contenant `SPEC.md`, les éventuels fichiers compagnons, et un `.decision-log.md`. Les appelants en mode headless reçoivent une réponse JSON avec le statut du résultat et la liste des fichiers écrits ou modifiés.
 
 :::note[Contrat de mutation]
 `bmad-spec` est le seul outil autorisé à écrire `SPEC.md` et les fichiers compagnons de la spécification. Les autres compétences produisent leurs propres artefacts natifs et invoquent `bmad-spec` en mode headless lorsqu’elles ont besoin d’exprimer une intention sous forme de contrat canonique ou de proposer des mises à jour.

docs/fr/reference/workflow-map.md

@@ -47,13 +47,13 @@ Définissez ce qu’il faut construire et pour qui.
 
 | Workflow   | Objectif                                                                                               | Livrable                                                                                                          |
 |------------|--------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------|
-| `bmad-prd` | Créez, mettez à jour ou validez un PRD[^1] — découverte accompagnée, trois intentions en un seul skill | Création/Mise à jour : `prd.md`, `addendum.md`, `.memlog.md` ; Validation : `validation-report.html` + `.md` |
+| `bmad-prd` | Créez, mettez à jour ou validez un PRD[^1] — découverte accompagnée, trois intentions en un seul skill | Création/Mise à jour : `prd.md`, `addendum.md`, `decision-log.md` ; Validation : `validation-report.html` + `.md` |
 | `bmad-ux`  | Concevez l’expérience utilisateur (lorsque l’UX compte)                                                | `DESIGN.md`, `EXPERIENCE.md`                                                                                      |
 
 :::tip[Trois intentions en un seul skill]
 `bmad-prd` couvre l’intégralité du cycle de vie du PRD. Précisez votre intention lors de l’appel, sinon le skill vous la demandera :
 
-- **Créer** — nouveau PRD à partir de zéro via une découverte accompagnée ; produit `prd.md`, `addendum.md` et `.memlog.md`
+- **Créer** — nouveau PRD à partir de zéro via une découverte accompagnée ; produit `prd.md`, `addendum.md` et `decision-log.md`
 - **Mettre à jour** — réconcilie un PRD existant avec un signal de changement, en mettant en évidence les conflits avant d’appliquer les modifications
 - **Valider** — évalue un PRD à l’aide d’une liste de contrôle configurable et produit un rapport de constats structuré au format HTML
 :::

docs/fr/tutorials/getting-started.md

@@ -147,7 +147,7 @@ Tous les workflows de cette phase sont optionnels. [**Vous ne savez pas lequel c
 **Pour les voies BMad Method et Enterprise :**
 
 1. Exécutez `bmad-prd` dans un nouveau chat — précisez votre intention (Create / Update / Validate) ou laissez le skill vous la demander
-2. Résultat : `prd.md`, `addendum.md`, `.memlog.md`
+2. Résultat : `prd.md`, `addendum.md`, `decision-log.md`
 
 :::note[Intentions de `bmad-prd`]
 

docs/how-to/customize-bmad.md

@@ -22,7 +22,7 @@ The `bmad-customize` skill is a guided authoring helper for the **per-skill agen
 :::note[Prerequisites]
 
 - BMad installed in your project (see [How to Install BMad](./install-bmad.md))
-- A way to run the resolver script — BMad is standardizing on `uv` (`uv run`, which provisions Python for you); a plain `python3` 3.11+ on your PATH still works during the transition. The script uses only stdlib `tomllib`, so there's nothing to `pip install`.
+- Python 3.11+ on your PATH (for the resolver script -- uses stdlib `tomllib`, no `pip install`, no `uv`, no virtualenv)
 - A text editor for TOML files
 :::
 
@@ -201,15 +201,15 @@ persistent_facts = [
 
 ## How Resolution Works
 
-On activation, the agent's SKILL.md runs a shared Python script that does the three-layer merge and returns the resolved block as JSON. The script uses only the Python standard library's `tomllib` module (no external dependencies). BMad is standardizing on `uv run` to invoke these scripts (uv provisions a suitable Python for you); a plain `python3` still works during the transition:
+On activation, the agent's SKILL.md runs a shared Python script that does the three-layer merge and returns the resolved block as JSON. The script uses the Python standard library's `tomllib` module (no external dependencies), so plain `python3` is enough:
 
 ```bash
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill {skill-root} \
   --key agent
 ```
 
-**Requirements**: Python 3.11+ (earlier versions don't include `tomllib`); nothing to `pip install`. Running via `uv run` is the going-forward standard — uv resolves a suitable interpreter for you. If you run it with `python3` directly during the transition, check your version with `python3 --version`: some platforms (macOS without Homebrew, Ubuntu 22.04) default `python3` to 3.10 or earlier, so you may need to install 3.11+ separately.
+**Requirements**: Python 3.11+ (earlier versions don't include `tomllib`). No `pip install`, no `uv`, no virtualenv. Check with `python3 --version`. Some platforms (macOS without Homebrew, Ubuntu 22.04) default `python3` to 3.10 or earlier, so you may need to install 3.11+ separately.
 
 `--skill` points at the skill's installed directory (where `customize.toml` lives). The skill name is derived from the directory's basename, and the script looks up `_bmad/custom/{skill-name}.toml` and `{skill-name}.user.toml` automatically.
 
@@ -217,17 +217,17 @@ Useful invocations:
 
 ```bash
 # Resolve the full agent block
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill /abs/path/to/bmad-agent-pm \
   --key agent
 
 # Resolve a single field
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill /abs/path/to/bmad-agent-pm \
   --key agent.icon
 
 # Full dump
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill /abs/path/to/bmad-agent-pm
 ```
 

docs/how-to/pressure-test-an-idea.md

@@ -1,55 +0,0 @@
----
-title: "Pressure-Test an Idea"
-description: Use the bmad-forge-idea skill to harden, prove, or kill an idea before you invest in it
-sidebar:
-  order: 12
----
-
-Use the `bmad-forge-idea` skill to put a half-formed idea under adversarial questioning. It either survives with earned conviction or dies cheaply.
-
-## When to Use This
-
-- You hold an idea and want it stress-tested before you commit time or money
-- You want an honest read on whether to kill it, not encouragement
-- You're choosing between branches of a decision and need each one resolved
-- Your idea lives inside an existing project and needs to be checked against what's already there
-
-## When to Skip This
-
-- You have no idea yet and need to generate options — use `bmad-brainstorming`
-- You've committed to a product and want it proven customer-first — use `bmad-prfaq`
-- You want your agents to debate a decision together — use `bmad-party-mode`
-
-:::note[Prerequisites]
-None. The forge runs in plain conversation. Installed agents and a configured persona roster make the session richer, but it works without them.
-:::
-
-## Run a Session
-
-### 1. Invoke the skill
-
-Type `bmad-forge-idea` in your IDE, or say "forge an idea" or "pressure-test this." Name the idea in the same message or wait for the first question.
-
-### 2. State your goal
-
-Tell the forge what you want: harden the idea, prove or kill it, or just think it through. The goal steers the questioning. Proving goes after the load-bearing claim first, and hardening drives each branch to a resolved answer.
-
-### 3. Defend your thinking, one branch at a time
-
-The interrogator asks one question at a time and puts its own recommended answer on the table for you to push against. Answer honestly. When it challenges a fuzzy term or a claim that doesn't match your project, settle that before you move on.
-
-### 4. Steer the room
-
-Every branch arrives with two voices — one from your roster, one conjured by the topic. Call a specific persona by name, summon a saved party, or say "adversarial on this" to have a claim attacked while you defend it.
-
-### 5. Land an exit
-
-Drive each branch to a resolved answer until the idea is hardened, killed, or simply clearer. Say when you're done, or let the forge call it.
-
-## What You Get
-
-The forge writes a self-contained `forge-report.html` every run, stamped to match the outcome. A hardened idea also distills into `forged-idea.md`, which captures the locked decisions and what was killed and why. That file feeds `bmad-spec`, `bmad-prd`, or `bmad-prfaq` for a product concept. A killed or clarified session needs no artifact; the report stands on its own.
-
-:::tip[Let it kill the idea]
-Finding out cheaply that an idea doesn't hold is the win. Don't steer the session toward a yes.
-:::

docs/reference/commands.md

@@ -99,8 +99,7 @@ Workflow skills run a structured, multi-step process without loading an agent pe
 | `bmad-create-epics-and-stories` | Create epics and stories |
 | `bmad-dev-story` | Implement a story |
 | `bmad-code-review` | Run a code review |
-| `bmad-quick-dev` | Implement a story or any other small intent — clarify, plan, implement, review, present |
-| `bmad-dev-auto` | One unattended development-loop iteration — small intent in, code out, no human interaction |
+| `bmad-quick-dev` | Unified quick flow — clarify intent, plan, implement, review, present |
 
 See [Workflow Map](./workflow-map.md) for the complete workflow reference organized by phase.
 

docs/reference/core-tools.md

@@ -18,7 +18,6 @@ Run any core tool by typing its skill name (e.g., `bmad-help`) in your IDE. No a
 | [`bmad-help`](#bmad-help) | Task | Get context-aware guidance on what to do next |
 | [`bmad-brainstorming`](#bmad-brainstorming) | Workflow | Facilitate interactive brainstorming sessions |
 | [`bmad-party-mode`](#bmad-party-mode) | Workflow | Orchestrate multi-agent group discussions |
-| [`bmad-forge-idea`](#bmad-forge-idea) | Workflow | Pressure-test an idea until it hardens, proves out, or dies cheaply |
 | [`bmad-spec`](#bmad-spec) | Workflow | Distill any intent input into a SPEC kernel and companions, the canonical contract for downstream work |
 | [`bmad-advanced-elicitation`](#bmad-advanced-elicitation) | Task | Push LLM output through iterative refinement methods |
 | [`bmad-review-adversarial-general`](#bmad-review-adversarial-general) | Task | Cynical review that finds what's missing and what's wrong |
@@ -71,7 +70,7 @@ Run any core tool by typing its skill name (e.g., `bmad-help`) in your IDE. No a
 
 **Input:** Brainstorming topic or problem statement, optional context file
 
-**Output:** a self-contained `brainstorm.html` keepsake of the session, an optional `brainstorm-intent.md` for downstream skills, and a `.memlog.md` session record
+**Output:** `brainstorming-session-{date}.md` with all generated ideas
 
 :::note[Quantity Target]
 The magic happens in ideas 50–100. The workflow encourages generating 100+ ideas before organization.
@@ -99,28 +98,6 @@ The magic happens in ideas 50–100. The workflow encourages generating 100+ ide
 
 **Output:** Real-time multi-agent conversation with maintained agent personalities
 
-## bmad-forge-idea
-
-**Pressure-test an idea until it hardens, proves out, or dies cheaply.** — An adversarial interrogator drives a half-formed idea one question at a time, bringing two characters to every branch, until what survives is something you can act on with conviction.
-
-**Use it when:**
-
-- You hold an idea and want it stress-tested before you invest in it
-- You want an honest read on whether to kill it
-- You need a thinking partner that pushes back instead of agreeing
-
-**How it works:**
-
-1. Establishes the goal up front and steers the questioning to match it
-2. Works one question at a time in dependency order, putting a recommended answer on the table to push against
-3. Brings two voices to every branch — one from your installed roster, one conjured by the topic
-4. Challenges fuzzy terms and tests claims against an existing project's material
-5. Lands as Hardened, Killed, or Clearer, with a self-contained report you can keep
-
-**Input:** The idea, in any domain — a feature, a business model, a research hypothesis, a life decision
-
-**Output:** A `forged-idea.md` distillate when an idea hardens (optional), plus a `forge-report.html` keepsake every run
-
 ## bmad-spec
 
 **Distill any intent input into the canonical SPEC contract for downstream work.** Takes a brief, PRD, GDD, RFC, brain dump, transcript, UX folder, or mixed multi-source input and produces a `SPEC.md` carrying the five-field kernel (Why, Capabilities, Constraints, Non-goals, Success signal) plus companion files for load-bearing content that does not fit the kernel.
@@ -136,7 +113,7 @@ The magic happens in ideas 50–100. The workflow encourages generating 100+ ide
 1. Reads the input and any ancillary linked materials.
 2. Distills into the five-field kernel using a configurable template; routes overflow into appropriately-named companions.
 3. Runs a two-pass self-validate (coherence rules, then preservation of every load-bearing source claim).
-4. Writes `SPEC.md`, sibling companions, and a `.memlog.md` under `{output_folder}/specs/spec-{slug}/`.
+4. Writes `SPEC.md`, sibling companions, and a `.decision-log.md` under `{output_folder}/specs/spec-{slug}/`.
 
 Spec Law enforces eight rules: capabilities carry both intent and success; intents are WHAT not HOW; constraints actually bend decisions; non-goals are explicit; success signals are concrete; capability IDs are stable; every load-bearing source claim is preserved; prose is lean.
 
@@ -146,7 +123,7 @@ Spec Law enforces eight rules: capabilities carry both intent and success; inten
 - `slug` (optional) — required only when input is sparse and no slug is derivable from a source filename.
 - `target_spec_path` (optional) — set to update an existing spec instead of creating a new one.
 
-**Output:** Spec folder containing `SPEC.md`, any companion files, and a `.memlog.md`. Headless callers receive a JSON response with the result status and the list of files written or modified.
+**Output:** Spec folder containing `SPEC.md`, any companion files, and a `.decision-log.md`. Headless callers receive a JSON response with the result status and the list of files written or modified.
 
 :::note[Mutation contract]
 `bmad-spec` is the only writer of `SPEC.md` and of spec-authored companions. Other skills produce their own native artifacts and invoke `bmad-spec` headless when they need to express intent as the canonical contract or propose updates.

docs/reference/workflow-map.md

@@ -35,11 +35,10 @@ it**](../explanation/analysis-phase.md).
 
 | Workflow                                                                  | Purpose                                                                    | Produces                  |
 |---------------------------------------------------------------------------|----------------------------------------------------------------------------|---------------------------|
-| `bmad-brainstorming`                                                      | Brainstorm Project Ideas with guided facilitation of a brainstorming coach | `brainstorm.html` keepsake plus an optional `brainstorm-intent.md` |
-| `bmad-forge-idea` | Pressure-test an idea until it hardens, proves out, or dies cheaply | `forge-report.html` every run; `forged-idea.md` when an idea hardens |
+| `bmad-brainstorming`                                                      | Brainstorm Project Ideas with guided facilitation of a brainstorming coach | `brainstorming-report.md` |
 | `bmad-domain-research`, `bmad-market-research`, `bmad-technical-research` | Validate market, technical, or domain assumptions                          | Research findings         |
-| `bmad-product-brief`                                                      | Capture strategic vision — best when your concept is clear                 | `brief.md` + `addendum.md`, plus any desired HTML or presentation output       |
-| `bmad-prfaq`                                                              | Working Backwards — stress-test your product concept customer-first             | `prfaq-{project}.md`      |
+| `bmad-product-brief`                                                      | Capture strategic vision — best when your concept is clear                 | `product-brief.md`        |
+| `bmad-prfaq`                                                              | Working Backwards — stress-test and forge your product concept             | `prfaq-{project}.md`      |
 
 ## Phase 2: Planning
 
@@ -47,13 +46,13 @@ Define what to build and for whom.
 
 | Workflow                | Purpose                                                                             | Produces                                          |
 |-------------------------|-------------------------------------------------------------------------------------|---------------------------------------------------|
-| `bmad-prd`              | Create, update, or validate a PRD — facilitated discovery, three intents in one skill | Create/Update: `prd.md`, `addendum.md`, `.memlog.md`; Validate: `validation-report.html` + `.md` |
-| `bmad-ux`               | Design user experience (when UX matters) — DESIGN.md (visual) + EXPERIENCE.md (behavioral) spine pair | `DESIGN.md`, `EXPERIENCE.md`, `.memlog.md`  |
+| `bmad-prd`              | Create, update, or validate a PRD — facilitated discovery, three intents in one skill | Create/Update: `prd.md`, `addendum.md`, `decision-log.md`; Validate: `validation-report.html` + `.md` |
+| `bmad-ux`               | Design user experience (when UX matters) — DESIGN.md (visual) + EXPERIENCE.md (behavioral) spine pair | `DESIGN.md`, `EXPERIENCE.md`, `.decision-log.md`  |
 
 :::tip[Three intents in one skill]
 `bmad-prd` handles the full PRD lifecycle. State your intent when invoking or the skill will ask:
 
-- **Create** — new PRD from scratch via coached discovery; produces `prd.md`, `addendum.md`, and `.memlog.md`
+- **Create** — new PRD from scratch via coached discovery; produces `prd.md`, `addendum.md`, and `decision-log.md`
 - **Update** — reconcile an existing PRD with a change signal, surfacing conflicts before applying changes
 - **Validate** — critique a PRD against a configurable checklist and produce a structured HTML findings report
 :::
@@ -68,13 +67,13 @@ Decide how to build it and break work into stories.
 
 | Workflow                              | Purpose                                    | Produces                    |
 |---------------------------------------|--------------------------------------------|-----------------------------|
-| `bmad-architecture`            | Make technical decisions explicit          | `ARCHITECTURE-SPINE.md` is the spine by default but can hydrate to your desired output or presentation needs also |
+| `bmad-create-architecture`            | Make technical decisions explicit          | `architecture.md` with ADRs |
 | `bmad-create-epics-and-stories`       | Break requirements into implementable work | Epic files with stories     |
 | `bmad-check-implementation-readiness` | Gate check before implementation           | PASS/CONCERNS/FAIL decision |
 
 ## Phase 4: Implementation
 
-Build it, one story at a time. Phase 4 epic and story automation is now available also. So you can choose how you want to stay in the loop. You can choose the full flow, or go right to quick flow.
+Build it, one story at a time. Coming soon, full phase 4 automation!
 
 | Workflow               | Purpose                                                                       | Produces                                             |
 |------------------------|-------------------------------------------------------------------------------|------------------------------------------------------|
@@ -94,7 +93,6 @@ Skip phases 1-3 for small, well-understood work.
 | Workflow         | Purpose                                                                   | Produces           |
 |------------------|---------------------------------------------------------------------------|--------------------|
 | `bmad-quick-dev` | Unified quick flow — clarify intent, plan, implement, review, and present | `spec-*.md` + code |
-| `bmad-dev-auto`  | One unattended development-loop iteration — small intent in, code out      | `spec-*.md` + code |
 
 ## Context Management
 

docs/tutorials/getting-started.md

@@ -70,10 +70,10 @@ BMad helps you build software through guided workflows with specialized AI agent
 
 | Phase | Name           | What Happens                                                 |
 | ----- | -------------- | ------------------------------------------------------------ |
-| 1     | Analysis       | Brainstorming, research, forge idea, product brief or PRFAQ _(optional)_ |
-| 2     | Planning       | Create requirements and design PRD, UX, SPEC                 |
-| 3     | Solutioning    | Design architecture spine or detailed project or system architectures          |
-| 4     | Implementation | Build epic by epic, story by story with quick dev or automated epic delivery    |
+| 1     | Analysis       | Brainstorming, research, product brief or PRFAQ _(optional)_ |
+| 2     | Planning       | Create requirements (PRD or spec)                            |
+| 3     | Solutioning    | Design architecture _(BMad Method/Enterprise only)_          |
+| 4     | Implementation | Build epic by epic, story by story                           |
 
 **[Open the Workflow Map](../reference/workflow-map.md)** to explore phases, workflows, and context management.
 
@@ -139,17 +139,16 @@ Create it manually at `_bmad-output/project-context.md` or generate it after arc
 All workflows in this phase are optional. [**Not sure which to use?**](../explanation/analysis-phase.md)
 
 - **brainstorming** (`bmad-brainstorming`) — Guided ideation
-- **forge-idea** (`bmad-forge-idea`) — Pressure-test an idea until it hardens or dies cheaply
 - **research** (`bmad-market-research` / `bmad-domain-research` / `bmad-technical-research`) — Market, domain, and technical research
 - **product-brief** (`bmad-product-brief`) — Recommended foundation document when your concept is clear
-- **prfaq** (`bmad-prfaq`) — Working Backwards challenge to stress-test your product concept customer-first
+- **prfaq** (`bmad-prfaq`) — Working Backwards challenge to stress-test and forge your product concept
 
 ### Phase 2: Planning (Required)
 
 **For BMad Method and Enterprise tracks:**
 
 1. Run `bmad-prd` in a new chat — state your intent (Create / Update / Validate) or let the skill ask
-2. Output: `prd.md`, `addendum.md`, `.memlog.md`
+2. Output: `prd.md`, `addendum.md`, `decision-log.md`
 
 :::note[`bmad-prd` intents]
 

docs/vi-vn/how-to/customize-bmad.md

@@ -22,7 +22,7 @@ Skill `bmad-customize` là trợ lý tạo cấu hình có hướng dẫn cho **
 :::note[Điều kiện tiên quyết]
 
 - BMad đã được cài trong dự án của bạn (xem [Cách cài đặt BMad](./install-bmad.md))
-- Một cách để chạy resolver script — BMad đang chuẩn hóa sang `uv` (`uv run`, tự cấp Python cho bạn); một `python3` 3.11+ thuần trên PATH vẫn dùng được trong giai đoạn chuyển đổi. Script chỉ dùng stdlib `tomllib`, nên không cần `pip install` gì cả.
+- Python 3.11+ có trên PATH của bạn (để chạy resolver; dùng stdlib `tomllib`, không cần `pip install`, `uv` hay virtualenv)
 - Một trình soạn thảo văn bản cho file TOML
 :::
 
@@ -201,15 +201,15 @@ persistent_facts = [
 
 ## Cách quá trình resolve diễn ra
 
-Khi agent được kích hoạt, `SKILL.md` của nó sẽ gọi một shared Python script để merge ba lớp nói trên và trả về block kết quả ở dạng JSON. Script này chỉ dùng `tomllib` của Python stdlib (không có dependency ngoài). BMad đang chuẩn hóa sang `uv run` để chạy các script này (uv tự cấp một bản Python phù hợp cho bạn); một `python3` thuần vẫn dùng được trong giai đoạn chuyển đổi:
+Khi agent được kích hoạt, `SKILL.md` của nó sẽ gọi một shared Python script để merge ba lớp nói trên và trả về block kết quả ở dạng JSON. Script này dùng `tomllib` của Python stdlib, nên `python3` thuần là đủ:
 
 ```bash
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill {skill-root} \
   --key agent
 ```
 
-**Yêu cầu**: Python 3.11+ vì các phiên bản cũ hơn không có `tomllib`; không cần `pip install` gì. Chạy qua `uv run` là chuẩn về sau — uv tự tìm một bản interpreter phù hợp cho bạn. Nếu bạn chạy trực tiếp bằng `python3` trong giai đoạn chuyển đổi, hãy kiểm tra phiên bản bằng `python3 --version`: trên một số nền tảng, `python3` mặc định vẫn là 3.10 hoặc thấp hơn, nên có thể bạn sẽ phải cài 3.11+ riêng.
+**Yêu cầu**: Python 3.11+ vì các phiên bản cũ hơn không có `tomllib`. Không cần `pip install`, không cần `uv`, không cần virtualenv. Bạn có thể kiểm tra bằng `python3 --version`. Trên một số nền tảng, `python3` mặc định vẫn là 3.10 hoặc thấp hơn, nên có thể bạn sẽ phải cài 3.11+ riêng.
 
 `--skill` trỏ vào thư mục skill đã cài, nơi có file `customize.toml`. Tên skill được lấy từ basename của thư mục, sau đó script sẽ tự tìm `_bmad/custom/{skill-name}.toml` và `{skill-name}.user.toml`.
 
@@ -217,17 +217,17 @@ Một số lệnh hữu ích:
 
 ```bash
 # Resolve toàn bộ block agent
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill /duong-dan/tuyet-doi/toi/bmad-agent-pm \
   --key agent
 
 # Resolve một trường cụ thể
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill /duong-dan/tuyet-doi/toi/bmad-agent-pm \
   --key agent.icon
 
 # Dump toàn bộ
-uv run {project-root}/_bmad/scripts/resolve_customization.py \
+python3 {project-root}/_bmad/scripts/resolve_customization.py \
   --skill /duong-dan/tuyet-doi/toi/bmad-agent-pm
 ```
 

evals/bmm-skills/bmad-product-brief/evals.json

@@ -0,0 +1,237 @@
+{
+  "skill_name": "bmad-product-brief",
+  "_design_notes": "Single-shot evals across two patterns. Pattern A (A1-A8) tests artifact correctness given complete inputs in headless mode. Pattern B tests process discipline (decision log fidelity, polish execution, intent boundaries) by inspecting transcript and side-artifacts. Facilitation/conversation-quality evals are deferred to a future multi-turn simulator.",
+  "evals": [
+    {
+      "id": "A1",
+      "_pattern": "artifact-correctness",
+      "prompt": "Run headless. Create a product brief for InsuLens.\n\nContext (use exactly this — do not invent):\n- Product: a smartphone app that pairs with off-the-shelf $200 thermal imaging accessories (FLIR ONE Pro and Seek Compact Pro). The app guides homeowners through a structured walkthrough and produces a professional-grade insulation audit in under 20 minutes.\n- Target: suburban homeowners aged 35-65 with houses built before 2000 (poor original insulation, rising energy bills).\n- Validation evidence: 50 user interviews completed in Q4 2025; 78% expressed willingness to pay $49 for a one-time audit if results were credible.\n- Stakes: this brief is the primary input investors will read before our first Series A pitch call.\n- Hardware dependency: requires a thermal imaging accessory (we do not manufacture hardware).\n- Known unknowns: insurance/warranty implications of homeowner-driven audits; whether the 78% intent translates to paid conversion at scale.\nRight-size for investor-stage rigor. Output a JSON status block at the end with status, intent, and artifact paths.",
+      "expected_output": "A run folder containing brief.md (with valid YAML frontmatter) and decision-log.md. Brief is 1-2 pages, addresses target audience, hardware dependency, validation evidence, and surfaces unknowns alongside knowns. Final assistant message includes JSON with status='complete', intent='create', and artifact paths.",
+      "files": [],
+      "expectations": [
+        "A run folder is created with brief.md and decision-log.md",
+        "brief.md has YAML frontmatter containing all four required fields: title, status, created (ISO 8601), updated (ISO 8601)",
+        "brief.md frontmatter status is 'draft' or 'final' (not missing or empty)",
+        "brief.md word count is between 250 and 1500",
+        "brief.md identifies the suburban-homeowner-aged-35-65 target audience",
+        "brief.md references the thermal imaging hardware dependency (FLIR ONE Pro / Seek Compact Pro or equivalent)",
+        "brief.md references the 50-user interview validation OR the 78% willingness-to-pay finding",
+        "brief.md surfaces at least one explicit unknown or unvalidated assumption",
+        "brief.md does not introduce facts absent from the input prompt (no invented competitors, fabricated metrics, or unstated partnerships)",
+        "The final assistant message contains a JSON object with keys 'status', 'intent', and at least one artifact path",
+        "The JSON 'intent' value is 'create'"
+      ]
+    },
+    {
+      "id": "A2",
+      "_pattern": "artifact-correctness",
+      "prompt": "Run headless. Create a brief for our app idea.",
+      "expected_output": "Headless mode with insufficient context should halt with status='blocked' and a reason field. No (or only skeleton) brief should be written. The skill must not invent a product to draft against.",
+      "files": [],
+      "expectations": [
+        "The final assistant message contains a JSON object with key 'status' set to 'blocked'",
+        "The JSON output includes a 'reason' field explaining what context is missing",
+        "If brief.md exists at all, its body content (excluding frontmatter) is fewer than 100 words OR is a placeholder skeleton — the skill did not invent a product",
+        "The transcript contains no clarifying questions sent back to the user (headless mode honored)"
+      ]
+    },
+    {
+      "id": "A3",
+      "_pattern": "artifact-correctness",
+      "prompt": "Run headless. Create a brief for our neighborhood compost coordinator app idea — we are moving forward with it. Q2 brainstorming session notes are at evals/bmm-skills/bmad-product-brief/files/q2-brainstorm.md; pull only what is relevant to the compost concept.",
+      "expected_output": "Brief focuses tightly on the compost coordinator concept. Source brainstorm is filtered, not ingested wholesale. Decision-log records that filtering occurred.",
+      "files": ["evals/bmm-skills/bmad-product-brief/files/q2-brainstorm.md"],
+      "expectations": [
+        "brief.md addresses the neighborhood compost coordinator concept",
+        "brief.md does not introduce content from unrelated brainstorm topics (weather + mood, meditation chime, podcasting tool, craft beer subscription, AI sommelier, office plants, ride coordinator, cookbook app, AR home staging)",
+        "brief.md word count is between 250 and 1500",
+        "brief.md incorporates at least 2 specific details from the compost section of the brainstorm (e.g., two-sided market with apartment dwellers and home compost-pile owners, hyperlocal neighborhood scope, free-at-launch with eventual subscription, Portland Sunnyside/Hawthorne pilot)",
+        "decision-log.md indicates the brainstorm was filtered for relevance, not ingested whole"
+      ]
+    },
+    {
+      "id": "A4",
+      "_pattern": "artifact-correctness",
+      "prompt": "Run headless. Validate the brief at evals/bmm-skills/bmad-product-brief/files/mossridge-brief/brief.md — the Mossridge Public Library board meets Monday and we need this to land. Read the addendum and decision-log in the same folder first. Cite specific sections, identify weaknesses, caveat what cannot be evaluated. Return inline only — no separate validation file.",
+      "expected_output": "Inline critique citing specific sections from the input brief. No new files. Caveats at least one claim that cannot be evaluated from the brief alone. Offers to roll findings into an Update.",
+      "files": [
+        "evals/bmm-skills/bmad-product-brief/files/mossridge-brief/brief.md",
+        "evals/bmm-skills/bmad-product-brief/files/mossridge-brief/addendum.md",
+        "evals/bmm-skills/bmad-product-brief/files/mossridge-brief/decision-log.md"
+      ],
+      "expectations": [
+        "The final output cites specific section names or line content from the input brief (not generic feedback)",
+        "The output identifies at least one specific weakness or area for improvement in the input brief",
+        "The output explicitly caveats at least one claim that cannot be evaluated from the brief alone (e.g., community demand, funding feasibility, volunteer sustainability)",
+        "The output offers to roll findings into an Update (or equivalent next-step proposal)",
+        "The final assistant message contains a JSON object with intent='validate'"
+      ]
+    },
+    {
+      "id": "A5",
+      "_pattern": "artifact-correctness",
+      "prompt": "Run headless. Create a brief for: a weekend-project iOS app called Sproutkeeper that reminds houseplant owners when to water their plants based on plant type and indoor humidity sensor data. Target is hobbyist plant owners. MVP scope only, single-developer side project, no investors, no team, just personal evening project.",
+      "expected_output": "Lightweight brief right-sized to a side project. Low rigor. No investor-grade framing.",
+      "files": [],
+      "expectations": [
+        "The final assistant message contains a JSON object with intent='create'",
+        "brief.md exists at the path referenced in the JSON output",
+        "brief.md is right-sized for a side project (closer to 250-500 words than 1500)",
+        "brief.md does not include investor-grade framing (no 'Series A inputs', 'TAM/SAM/SOM', 'go-to-market strategy' boilerplate when the user said this is a personal evening project)",
+        "The transcript contains no clarifying questions to the user",
+        "Sections that do not earn their place for a side project are dropped or kept minimal (e.g., no extensive Risk or Success Criteria padding)"
+      ]
+    },
+    {
+      "id": "A6",
+      "_pattern": "artifact-correctness",
+      "prompt": "Run headless. Create a brief from this memo. It is from our last working group on a new microcredential program at Branfield Community College. Memo is at evals/bmm-skills/bmad-product-brief/files/branfield-memo.md. Use what is there; do not re-elicit facts already present.",
+      "expected_output": "Brief reflects content from the memo. No re-asking for facts already present. Decision-log notes ingestion of the memo.",
+      "files": ["evals/bmm-skills/bmad-product-brief/files/branfield-memo.md"],
+      "expectations": [
+        "brief.md incorporates at least 3 distinct facts or decisions present in the input memo",
+        "decision-log.md references having used the memo as source material",
+        "The transcript does not ask the user to re-state the program name, target student, or core curriculum focus if those are present in the memo",
+        "brief.md does not invent program details not present in the memo"
+      ]
+    },
+    {
+      "id": "A7",
+      "_pattern": "artifact-correctness",
+      "prompt": "Run headless. Create a brief for Brightway — our smart bike helmet with crash detection, turn signals, and braking lights. Meridian Insights produced a market research report on e-mobility at evals/bmm-skills/bmad-product-brief/files/meridian-mobility-report.md. Use only what is relevant to the safety helmet category — do not let the e-scooter, charging-infrastructure, or bike-share segments bleed into the brief.",
+      "expected_output": "Brief focuses on the smart bike helmet concept. Pulls relevant findings from the helmet section. Other mobility segments do not appear.",
+      "files": ["evals/bmm-skills/bmad-product-brief/files/meridian-mobility-report.md"],
+      "expectations": [
+        "brief.md addresses the Brightway smart bike helmet concept",
+        "brief.md does not introduce content from unrelated mobility segments (e-scooters, charging infrastructure, bike-share, vehicle-to-grid)",
+        "brief.md word count is between 250 and 1500",
+        "brief.md incorporates at least 2 specific findings from the smart helmet section of the report (e.g., market sizing, key players, crash detection technology trends, regulatory or insurance landscape)",
+        "decision-log.md indicates the report was filtered to the helmet category rather than ingested whole"
+      ]
+    },
+    {
+      "id": "A8",
+      "_pattern": "artifact-correctness",
+      "prompt": "Run headless. Create a brief for Pantry Bridge — a meal-kit subscription targeted at adults 65+ who live alone and want fresh meals without grocery shopping. Customer research transcripts are at evals/bmm-skills/bmad-product-brief/files/pantry-bridge-interviews.md. Pull what is relevant from the older-adult interviews; do not conflate insights from the working-parent, student, or corporate-buyer personas.",
+      "expected_output": "Brief focuses on the older-adult target persona. Eleanor's interview drives the insights. Other personas do not pollute the brief.",
+      "files": ["evals/bmm-skills/bmad-product-brief/files/pantry-bridge-interviews.md"],
+      "expectations": [
+        "brief.md addresses the Pantry Bridge older-adult meal-kit concept",
+        "brief.md does not conflate insights from non-target personas (working parent Susan, college student Marcus, corporate cafeteria buyer Dimitri)",
+        "brief.md word count is between 250 and 1500",
+        "brief.md incorporates at least 2 specific insights from Eleanor's interview (e.g., grocery-trip difficulty, portion sizing, dietary restrictions, social aspects of meals, trust concerns)",
+        "decision-log.md notes which interviews were used and which were excluded"
+      ]
+    },
+    {
+      "id": "B1",
+      "_pattern": "process-discipline",
+      "prompt": "Run headless. Create a brief for HelmStack — an open-source observability platform for distributed systems.\n\nWe have made these specific decisions and want each captured in the decision log with rationale:\n\n1. Pricing: Free open-source core; paid SaaS at $29/seat/month. Rejected paid-one-shot-license model because it would limit network effects in the OSS community.\n2. Launch: Invite-only beta for 6 weeks before public launch. Rejected open public launch — operational risk too high before stability is proven on real workloads.\n3. Stack: TypeScript + Postgres for the backend. Rejected Go + MongoDB — TypeScript aligned better with our team's existing skills and the frontend codebase.\n4. ICP: 5-50 person engineering teams for MVP. Rejected enterprise-first focus because the sales cycle is too long for our capital runway.\n5. Self-host: SaaS-only at launch; self-host arrives in v2. Rejected concurrent self-host because it would slow shipping velocity past our funding window.\n\nProduce brief.md and decision-log.md.",
+      "expected_output": "Decision log contains all five named decisions with rationale captured. Brief reflects the decisions but the decision log is the canonical record.",
+      "files": [],
+      "expectations": [
+        "decision-log.md exists in the run folder",
+        "decision-log.md captures the pricing decision (free OSS + $29/seat SaaS) with the rejected alternative (paid one-shot license) and rationale (network effects)",
+        "decision-log.md captures the invite-only-beta decision with the rejected alternative (open public launch) and rationale (operational risk before stability)",
+        "decision-log.md captures the platform-stack decision (TypeScript + Postgres) with the rejected alternative (Go + MongoDB) and rationale (team skills / frontend alignment)",
+        "decision-log.md captures the ICP decision (5-50 person eng teams) with rationale referencing sales cycle / runway",
+        "decision-log.md captures the self-host-timing decision (SaaS-only at launch, self-host v2) with rationale (shipping velocity / funding window)"
+      ]
+    },
+    {
+      "id": "B2",
+      "_pattern": "process-discipline",
+      "prompt": "Run headless. Create a brief for HelmStack — an open-source observability platform for distributed systems.\n\nWe have made these specific decisions and want each captured in the decision log with rationale:\n\n1. Pricing: Free open-source core; paid SaaS at $29/seat/month. Rejected paid-one-shot-license model because it would limit network effects in the OSS community.\n2. Launch: Invite-only beta for 6 weeks before public launch. Rejected open public launch — operational risk too high before stability is proven on real workloads.\n3. Stack: TypeScript + Postgres for the backend. Rejected Go + MongoDB — TypeScript aligned better with our team's existing skills and the frontend codebase.\n4. ICP: 5-50 person engineering teams for MVP. Rejected enterprise-first focus because the sales cycle is too long for our capital runway.\n5. Self-host: SaaS-only at launch; self-host arrives in v2. Rejected concurrent self-host because it would slow shipping velocity past our funding window.\n\nProduce brief.md and decision-log.md.",
+      "expected_output": "Brief is consistent with the decision log: every decision in the log is reflected in the brief, and no claim in the brief is absent from the input prompt or the log. Tests bidirectional fidelity.",
+      "files": [],
+      "expectations": [
+        "brief.md mentions the OSS-core + paid-SaaS pricing structure",
+        "brief.md references the invite-only-beta launch sequencing OR identifies the launch model consistent with the decision log",
+        "brief.md references the platform-stack choice (TypeScript + Postgres) OR is silent on stack — but does not contradict it (no mention of Go, MongoDB, etc.)",
+        "brief.md identifies 5-50 person eng teams as the ICP (or equivalent — small-to-mid-size eng teams)",
+        "brief.md does not introduce decisions, competitors, partnerships, metrics, or product features absent from both the input prompt and decision-log.md (no invented facts)",
+        "Each substantive decision in decision-log.md has a corresponding reflection in brief.md (no log-to-brief drops)"
+      ]
+    },
+    {
+      "id": "B3",
+      "_pattern": "process-discipline",
+      "prompt": "Run headless. Create a product brief for InsuLens.\n\nContext (use exactly this — do not invent):\n- Product: a smartphone app that pairs with off-the-shelf $200 thermal imaging accessories (FLIR ONE Pro and Seek Compact Pro). The app guides homeowners through a structured walkthrough and produces a professional-grade insulation audit in under 20 minutes.\n- Target: suburban homeowners aged 35-65 with houses built before 2000.\n- Validation: 50 user interviews completed in Q4 2025; 78% willingness to pay $49 for a one-time audit.\n- Stakes: Series A pitch input.\n- Hardware: requires a thermal accessory (we do not manufacture hardware).\n\nProduce brief.md and decision-log.md. Run the polish phase before presenting.",
+      "expected_output": "The transcript shows the polish phase executing — the skill invokes bmad-editorial-review-structure and bmad-editorial-review-prose, either via the Skill tool directly or via Agent tool calls whose description or prompt targets those editorial skills. Both passes must occur after the initial draft is written and before the final JSON status block.",
+      "files": [],
+      "expectations": [
+        "The transcript contains either a Skill tool call invoking bmad-editorial-review-structure, OR an Agent tool call whose description or prompt references structural review or bmad-editorial-review-structure",
+        "The transcript contains either a Skill tool call invoking bmad-editorial-review-prose, OR an Agent tool call whose description or prompt references prose review or bmad-editorial-review-prose",
+        "Both editorial-pass dispatches (Skill or Agent) occur after the first Write tool call that creates brief.md",
+        "Both editorial-pass dispatches (Skill or Agent) occur before the final assistant message containing the JSON status block"
+      ]
+    },
+    {
+      "id": "B5",
+      "_pattern": "process-discipline",
+      "prompt": "Run headless. Update the brief at evals/bmm-skills/bmad-product-brief/files/forkbird-brief/brief.md — we have decided to add B2B catering services for corporate events, in addition to the direct-to-consumer delivery model. Read the existing decision-log.md and addendum.md in the same folder first.",
+      "expected_output": "The skill MUST detect the contradiction with the prior 'rejected B2B catering for MVP' decision (in decision-log.md) before applying the change. Acceptable resolutions: (a) halt with blocked status surfacing the conflict, or (b) apply the change with addendum.md capturing the override and rationale. Brief must not silently flip without acknowledging the prior decision.",
+      "files": [
+        "evals/bmm-skills/bmad-product-brief/files/forkbird-brief/brief.md",
+        "evals/bmm-skills/bmad-product-brief/files/forkbird-brief/addendum.md",
+        "evals/bmm-skills/bmad-product-brief/files/forkbird-brief/decision-log.md"
+      ],
+      "expectations": [
+        "The transcript or output explicitly references the prior 'rejected B2B catering for MVP' decision from decision-log.md",
+        "The contradiction is surfaced before the brief body is modified (a Read of decision-log.md occurs before the Edit/Write to brief.md, AND the conflict is named in the assistant output)",
+        "Either the JSON status is 'blocked' with the conflict in the reason field, OR addendum.md is updated with an override entry capturing the rationale for reversing the prior decision",
+        "If the brief is updated, decision-log.md gains a new entry referencing the catering reversal",
+        "If the brief is updated, the YAML frontmatter 'updated' field is later than the original 'created' field"
+      ]
+    },
+    {
+      "id": "B6",
+      "_pattern": "process-discipline",
+      "prompt": "Run headless. Update the brief at evals/bmm-skills/bmad-product-brief/files/forkbird-brief/brief.md — we have signed our fifth chef partner (Chicago metro). Add this to the existing operating-model and what's-known sections. Read the existing decision-log.md first.",
+      "expected_output": "Clean update — does not contradict any prior decision. Brief gets updated, decision-log gains a new entry, YAML 'updated' bumps but 'created' stays the same. No spurious addendum since this is a status update, not an override.",
+      "files": [
+        "evals/bmm-skills/bmad-product-brief/files/forkbird-brief/brief.md",
+        "evals/bmm-skills/bmad-product-brief/files/forkbird-brief/addendum.md",
+        "evals/bmm-skills/bmad-product-brief/files/forkbird-brief/decision-log.md"
+      ],
+      "expectations": [
+        "brief.md is updated to reflect the signed fifth chef partner in Chicago",
+        "brief.md frontmatter 'updated' field is later than the original 'created' timestamp; 'created' is unchanged",
+        "decision-log.md contains a new entry referencing the fifth chef signing",
+        "The transcript does not surface a fictional contradiction — this is a clean update, not an override of a prior decision"
+      ]
+    },
+    {
+      "id": "B7",
+      "_pattern": "process-discipline",
+      "prompt": "Run headless. Validate the brief at evals/bmm-skills/bmad-product-brief/files/mossridge-brief/brief.md — we are presenting to the library board Monday. Read the addendum and decision-log in the same folder. Cite specific sections. Return inline only.",
+      "expected_output": "Validate is read-only. No new files created. No existing files modified. Critique returned inline in the assistant output.",
+      "files": [
+        "evals/bmm-skills/bmad-product-brief/files/mossridge-brief/brief.md",
+        "evals/bmm-skills/bmad-product-brief/files/mossridge-brief/addendum.md",
+        "evals/bmm-skills/bmad-product-brief/files/mossridge-brief/decision-log.md"
+      ],
+      "expectations": [
+        "No new files appear in the mossridge-brief artifacts directory after the run (only the three input files)",
+        "The input brief.md, addendum.md, and decision-log.md are byte-identical to the staged fixtures (no Edit/Write tool calls modified them)",
+        "The transcript contains no Write tool calls and no Edit tool calls targeting the mossridge-brief folder",
+        "The final assistant message contains a JSON object with intent='validate'"
+      ]
+    },
+    {
+      "id": "C1",
+      "_pattern": "config-compliance",
+      "prompt": "Run headless. Create a product brief for TaskFlow — a lightweight daily planning app for freelancers who juggle multiple clients. Core idea: a single daily view that pulls together tasks, time blocks, and client context so the freelancer always knows what to work on next. Target is independent freelancers, 1-3 clients at a time, who currently manage their day across sticky notes, calendar apps, and spreadsheets. MVP is mobile-first. No investors — the founder is bootstrapping.",
+      "expected_output": "Brief written in Spanish (document_output_language=Spanish). Assistant's conversational output reflects the configured British-accent communication style. Brief lands at the custom output path (test-output/artifacts/briefs/...) rather than the default _bmad-output path. Brief is right-sized for a bootstrapped solo project.",
+      "files": [],
+      "expectations": [
+        "brief.md exists under test-output/artifacts/briefs/ (the custom planning_artifacts path), not under _bmad-output/",
+        "The final JSON status block artifact paths reference test-output/ rather than _bmad-output/",
+        "brief.md body is written in Spanish — the majority of prose content (headings, section bodies) is in Spanish, not English",
+        "brief.md covers the TaskFlow concept: freelancer daily planning, multi-client context, the sticky-notes-plus-calendar-plus-spreadsheet problem",
+        "brief.md is right-sized for a bootstrapped side project — appropriate depth and scope for a solo-founder app with no investor audience, no TAM/SAM/SOM framing, no Series A language, and no sections that pad for enterprise credibility",
+        "The assistant's non-document output (transcript text content outside of brief.md) contains at least one marker of British informal register (e.g., 'mate', 'cheers', 'brilliant', 'sorted', 'innit', 'blimey', 'proper', 'right then', or equivalent pub-idiom phrasing)"
+      ]
+    }
+  ]
+}

evals/bmm-skills/bmad-product-brief/files/branfield-memo.md

@@ -0,0 +1,46 @@
+# Working Group Notes — Microcredential Program
+
+**Branfield Community College**
+**Meeting:** 2026-04-22
+**Attendees:** Provost, Workforce Dev Director, Chair of Industry Advisory Board, two faculty leads (Data Analytics, Healthcare Admin), Financial Aid Director
+
+## Why we're doing this
+
+Regional employer survey (Q1 2026) showed 340+ unfilled mid-skill jobs in the three-county area. State workforce board approved a $1.4M grant if we can launch by fall 2027 with at least three tracks. Existing AAS programs are too long for working adults — average completion 3.5 years.
+
+## What we're building
+
+Six-month stackable microcredentials. Three tracks at launch:
+
+1. **Data Analytics** (SQL, Excel/Power BI, intro Python). Faculty lead Marisol Reyes. Strongest employer demand. Will be MVP — first to launch, used to validate format.
+2. **Healthcare Admin** (medical coding, EHR systems, patient workflow). Faculty lead Dev Patel. Aging population in region drives demand.
+3. **Sustainable Construction** (green building practices, retrofit basics, code compliance). New faculty hire required.
+
+Stackable means credits transfer into related AAS or BAS later if the student wants.
+
+## Decisions made today
+
+- **Data Analytics is MVP.** Launch fall 2027, others phase in spring/fall 2028. Validate format before scaling.
+- **Hybrid delivery.** Two evenings/week in person + asynchronous online. Board rejected pure-online (concerns about adult learner outcomes data).
+- **Stipend program.** Up to $3,000/student for low-income students, funded from the state grant. Means-tested.
+- **Industry Advisory Board** has approval authority on curriculum. Three employers committed (regional hospital, mid-size data consultancy, county housing authority). All three commit to interview every graduate.
+- **Cohort cap: 24 per track per term.** Driven by classroom size and faculty load.
+
+## Open questions
+
+- Childcare for evening sessions — can we partner with the campus childcare center? Deferred to next meeting.
+- Marketing — provost wants to know cost per enrolled student before approving budget. Need workforce dev to model.
+- Do we offer a tuition payment plan in addition to the stipend? Financial aid director thinks yes; provost wants to see uptake projections first.
+
+## What we're NOT doing
+
+- Not pursuing pure-online delivery (rejected — see above).
+- Not launching all three tracks at once (rejected — risk concentration, faculty bandwidth).
+- Not building employer-customized cohorts (rejected — too operationally complex for MVP).
+
+## Next steps
+
+- Workforce Dev: marketing cost model by 2026-05-15.
+- Provost: childcare partnership exploratory conversation.
+- Faculty leads: draft data analytics curriculum outline by 2026-06-01.
+- Reconvene 2026-05-20.

evals/bmm-skills/bmad-product-brief/files/forkbird-brief/addendum.md

@@ -0,0 +1,40 @@
+# Addendum — Forkbird Kitchen
+
+## Options considered (and not taken)
+
+### B2B / corporate catering
+
+Considered as a parallel revenue stream from day one. Rejected for MVP. Different operational rhythm (bulk orders, fixed delivery windows, invoiced billing), different customer (procurement, not eaters), different unit economics. Splitting attention at launch risked degrading both. Revisit if consumer foundation is established by month 12.
+
+### Subscription / meal plan
+
+Considered as a recurring-revenue layer. Rejected for MVP. Operationally expensive at our planned scale: requires demand forecasting per subscriber, kitchen scheduling locked further out, and packaging/refrigerated handling we are not yet equipped for. Reasonable to revisit once kitchen utilization stabilizes.
+
+### Retail / grocery channel
+
+Considered (refrigerated meals in Whole Foods, Sprouts). Rejected for MVP. Different product (cold meals, longer shelf life, different texture profile), different go-to-market (broker relationships, slotting fees, category management). Parked for year 2 — would require a separate product line, not a channel extension.
+
+### Lower-priced everyday tier
+
+Considered. Rejected for now. The brand position is chef-driven; introducing a value tier alongside risks the premium signal in marketplace search ranking and review patterns. Explored alternative of separate brand for value tier; deferred.
+
+## Personas (extended)
+
+**The plant-based weekday professional.** Lives in a dense urban neighborhood, orders 4–6 times a month, splits between own-cooking and delivery. Sources of dissatisfaction with current options: chain plant-based menus feel formulaic, fine-dining plant-based is too expensive for weeknight, marketplace search surfaces too many low-quality options.
+
+**The dietary-flex household member.** One person in a household is plant-based by preference; the other(s) are not. Ordering pattern is "tonight one of us wants Forkbird, the other wants something else." We benefit from being a dependable single-cuisine option that doesn't require negotiating across diets.
+
+## Sizing notes
+
+- Total addressable: ~6.2M urban professionals across 5 metros eating plant-based 3+ times/week (based on 2024 Plant Based Foods Association data, urban segmentation).
+- Serviceable addressable (within delivery radius of planned kitchens at launch): ~840K.
+- Realistic Y1 capture (per metro forecast): 0.4% of SAM = 3,360 active customers across all metros.
+
+## Sourcing standard — exact wording
+
+"For each dish on the menu, we publish the source of every ingredient that represents at least 5% of cost. We commit that at least 60% of total ingredient weight is sourced within 200 miles of the kitchen preparing that dish. Both numbers are auditable; we publish them per-dish in the app. If we cannot meet the 60% local threshold for a dish, the dish does not ship."
+
+## Technical constraints
+
+- Marketplace integration (DoorDash, UberEats, Grubhub) requires their menu management API. We are using a third-party middleware (Olo) to avoid maintaining three separate integrations.
+- Ingredient transparency display requires structured data per dish. We need an ingredient-master database; current option is to extend our recipe-management software vendor.

evals/bmm-skills/bmad-product-brief/files/forkbird-brief/brief.md

@@ -0,0 +1,56 @@
+---
+title: Forkbird Kitchen — Product Brief
+status: final
+created: 2026-02-14
+updated: 2026-02-14
+---
+
+# Forkbird Kitchen
+
+## What it is
+
+A delivery-only ghost kitchen brand offering chef-driven plant-based meals in five US metros: San Francisco, New York, Los Angeles, Seattle, and Chicago. Launch operating model is direct-to-consumer through our own iOS/Android app and the major third-party marketplaces (DoorDash, UberEats, Grubhub).
+
+## Who it's for
+
+Urban professionals aged 28–45 who eat plant-based meals at least three times a week, value chef-driven food over chain alternatives, and order delivery 4+ times monthly. Initial geographic focus is dense neighborhoods within 3-mile delivery radii of partner kitchens.
+
+We are not building for: families with children (different ticket size and ordering pattern), occasional plant-based eaters (price sensitivity too high for our positioning), or office lunch (different time-of-day operation).
+
+## Why it wins
+
+Three things are deliberately stacked:
+
+1. **Chef partnerships, not chef-as-marketing.** Each metro has a named chef (with prior fine-dining or notable plant-based credit) who designs the rotating menu and earns equity in that metro's P&L. They are not endorsers; they are operators.
+2. **Ingredient sourcing standards.** Published per-dish: where it came from, how it was farmed, what portion of cost it represents. No dish ships if we can't source within 200 miles for ≥60% of ingredient weight. This is auditable, not marketing copy.
+3. **Speed without cars.** Average ticket-to-door is 28 minutes from order placement, achieved by tight delivery radii and dense order density per kitchen. Long delivery erodes plant-based texture more than animal protein — speed is product, not logistics.
+
+## Operating model
+
+Five kitchens, one per metro, each leased space inside an existing food-prep facility. No customer-facing storefronts. App orders go through our stack; marketplace orders pass through their stacks. Menu rotates every six weeks per chef.
+
+Pricing tier: $14–$22 per entrée before delivery. We are deliberately at chef-driven positioning, not value positioning.
+
+## What's known
+
+- Demand validated through three pop-up dinners in SF and NY (Q4 2025). 480 covers, 78% repeat intent based on post-event survey.
+- Operating partner identified in each metro. Leases signed for SF, NY, LA. Seattle and Chicago in negotiation.
+- Three of five chefs signed; two in active conversations.
+
+## What's unknown
+
+- Whether ingredient-sourcing transparency is a differentiator at point of sale (in-app) or only in marketing. Our hypothesis is "both" but we have not tested in-app.
+- Marketplace economics. DoorDash takes 15–30% depending on tier; we are modeling the lower tier but have not negotiated.
+- Whether the 3-mile radius holds outside SF/NY (lower density in LA/Chicago).
+
+## Risks
+
+- Chef churn. If a metro chef leaves, the metro brand loses its anchor. Mitigation: equity vesting over 24 months, named-chef terms in operating agreement.
+- Sourcing cost volatility. 60% local-within-200-miles can spike with weather/supply disruption. We have not modeled the worst case.
+- Marketplace dependency. If DoorDash terms shift adversely, our blended margin is at risk. We are deliberately building the owned-app channel to reduce this dependency.
+
+## Success criteria for first 12 months
+
+- 4 of 5 metros operating profitably at the unit level (kitchen + chef + delivery economics) by month 9
+- 30% of orders through owned app (vs. marketplaces) by month 12
+- Chef retention 100% through year 1

evals/bmm-skills/bmad-product-brief/files/forkbird-brief/decision-log.md

@@ -0,0 +1,27 @@
+# Decision Log — Forkbird Kitchen
+
+## 2026-01-08
+- **Brand position: chef-driven, premium plant-based.** Considered value tier; rejected for MVP. Premium positioning is the wedge against marketplace generic plant-based.
+
+## 2026-01-12
+- **Five-metro launch: SF, NY, LA, Seattle, Chicago.** Considered three-metro start; rejected as not enough density to test the chef-equity model meaningfully.
+- **Ghost kitchen, no storefront.** Storefronts ruled out — capex too high for MVP, dilutes the speed advantage.
+
+## 2026-01-19
+- **Pricing tier $14–$22 per entrée.** Modeled against three competitor sets: chain plant-based, fine-dining plant-based delivery, generic mid-tier delivery. Sits cleanly above chain, below fine-dining.
+- **Chef equity in metro P&L.** Rejected flat fee + revenue share alternative; equity creates the operator incentive we want.
+
+## 2026-01-26
+- **Rejected B2B catering segment for MVP.** Different operational rhythm and customer; would split attention at launch and risk degrading both consumer and B2B execution. Revisit in year 2 if consumer foundation is solid. (Discussion: 2 hours; chef partners weighed in against splitting focus; CFO modeled the dilution effect on consumer kitchen utilization.)
+- **Rejected subscription model for MVP.** Operationally expensive at planned scale; revisit once kitchen utilization stabilizes.
+
+## 2026-02-02
+- **Sourcing standard: 60% within 200 miles, published per-dish.** Considered weaker thresholds (50% / 250 miles); rejected as not differentiating enough to be worth publishing. The number has to be defensible.
+- **Marketplace channel mix: own app + DoorDash + UberEats + Grubhub.** Considered own-app only; rejected as too slow on demand acquisition. Considered marketplaces only; rejected — own app is critical to long-term margin.
+
+## 2026-02-09
+- **Six-week menu rotation per chef.** Considered four-week (more freshness) and eight-week (more operational stability). Six is the compromise; reassess after first two cycles.
+- **Marketing budget: 60% acquisition / 40% brand.** Rejected pure-acquisition because chef-driven positioning needs brand-level signal that paid acquisition alone won't carry.
+
+## 2026-02-14
+- **Brief finalized for Series A inputs.** Status moved to final.

evals/bmm-skills/bmad-product-brief/files/meridian-mobility-report.md

@@ -0,0 +1,116 @@
+# E-Mobility Market Report 2026
+
+**Prepared by:** Meridian Insights
+**Date:** Q2 2026
+**Coverage:** North America, with comparative reference to EU markets
+**Engagement code:** MI-2026-EMOB-007
+
+---
+
+## Executive Summary
+
+The e-mobility category continues a multi-year structural shift from "alternative transportation" to mainstream mobility infrastructure. North American unit volume across e-bikes, e-scooters, and connected safety hardware grew 18% year-over-year in 2025, against a 6% growth rate for traditional bicycles. Three macro factors are durably reshaping the category: regulatory clarity at the state level (29 US states now have explicit e-bike classifications, up from 14 in 2022), insurance industry interest in telematics-style risk pricing, and a generational shift in commuting preferences among the 28-44 cohort.
+
+This report covers seven segments of the broader e-mobility landscape: e-bike retail, e-scooter regulation, bike-share systems, charging infrastructure, smart helmet hardware, and grid-integration trends. Findings are synthesized from 142 stakeholder interviews, 18 retailer site visits, government regulatory filings, and proprietary point-of-sale data from 4,200 specialty retail outlets.
+
+---
+
+## Methodology
+
+Quantitative data was sourced from Meridian's proprietary Mobility Retail Panel (MRP), which aggregates POS data from independent specialty retailers and select chain operators. Where panel data is incomplete or lagging, we supplemented with manufacturer-reported shipment volumes and customs/import filings. Qualitative findings draw on 142 interviews conducted between November 2025 and March 2026 with retailers, fleet operators, regulators, manufacturers, and end users.
+
+Helmet category sizing uses a separate methodology described in Section 8, blending CPSC compliance filings, manufacturer disclosures, and a sample purchase-intent survey of 3,400 cyclists.
+
+---
+
+## Section 3: Market Sizing — Total E-Mobility
+
+The North American e-mobility market reached an estimated $14.7B in retail volume in 2025, up from $12.5B in 2024. The largest segment by volume is e-bikes at $7.2B, followed by e-scooter retail at $2.8B (excluding shared-fleet operations), bike-share and dockless mobility services at $2.1B, charging infrastructure at $1.8B, and connected safety hardware at $0.8B.
+
+Compound annual growth rate (CAGR) forecasts through 2030 vary substantially by segment. We forecast 14% CAGR for e-bikes, 6% for e-scooters (decelerating as the regulatory regime stabilizes), 9% for bike-share, 22% for charging infrastructure (driven by both bike and scooter charging), and 31% for connected safety hardware (off a smaller base). Vehicle-to-grid (V2G) integration is too early to forecast reliably; we treat it as an emerging segment.
+
+---
+
+## Section 4: E-Bike Market Deep Dive
+
+E-bikes represent the largest single segment by retail value. The 2025 unit mix favored Class 1 (pedal-assist, max assisted speed 20 mph) at 58% of units, Class 2 (throttle, max 20 mph) at 24%, and Class 3 (pedal-assist, max 28 mph) at 18%. Class 3 is the fastest-growing classification on a unit basis, driven by suburban commuter demand.
+
+Manufacturer concentration shifted in 2025. The top 10 brands by unit volume now hold 64% of the market, up from 51% in 2022 — consolidation that mirrors patterns seen in the traditional bicycle market in the early 2000s. Specialized, Trek, and Cannondale (operating their respective electric sub-brands) represent the top three. Direct-to-consumer brands (Rad Power, Lectric, Aventon) collectively hold approximately 19% of retail value.
+
+Retail channel split favored independent specialty bike shops at 47% of unit volume, with direct-to-consumer at 28%, big-box retail at 17%, and e-commerce marketplaces (Amazon, Walmart.com) at 8%. The independent specialty channel commands a price premium of approximately 22% over comparable D2C alternatives, attributed to in-store fitting, post-sale service relationships, and higher-margin component upgrades.
+
+Notable trends in 2025: cargo e-bike sub-segment grew 41% YoY (small base, dense urban geographies); battery range claims continue to drift upward with manufacturer claims of 60+ mile range becoming standard for $2,500+ price points; bottom-bracket motor placement (mid-drive) gained share over hub-drive in the $3,000+ tier.
+
+---
+
+## Section 5: E-Scooter Regulatory Landscape
+
+The North American e-scooter regulatory environment matured significantly during 2024-2025 after several years of municipal experimentation and reactive policymaking. Forty-one US cities now operate under what we classify as "stable" regulatory regimes (defined as: explicit operating permit framework, defined sidewalk/bike-lane rules, helmet provisions, and revenue-share or fee structures with the city). This is up from 19 cities in 2022.
+
+The regulatory shift has compressed operator margins. Permit fees and per-trip surcharges in major markets (Los Angeles, Chicago, Atlanta, Denver) range from $0.15 to $0.42 per trip, against average ride revenue of $5.40. Several major operators have exited markets where permit economics have proven unviable; Lime exited five secondary US markets in 2025 citing exactly this reason.
+
+Helmet requirements remain inconsistent. Thirteen US states require helmets for riders under 18 only; seven require them for all riders; the rest leave it to municipalities. Enforcement is widely acknowledged to be minimal even where mandates exist. EU markets are substantially stricter, with mandatory helmet provisions in France, Germany, and Italy applying to all e-scooter riders.
+
+Insurance treatment is also fragmenting. Five US states have classified e-scooters as "motor vehicles" requiring liability coverage, raising the floor on operating costs for shared-fleet providers. Most states still treat them as bicycles for insurance purposes.
+
+---
+
+## Section 6: Bike-Share and Dockless Mobility
+
+Docked bike-share systems (Citi Bike, Divvy, Bluebikes, Capital Bikeshare) continue stable, slow growth. Capital Bikeshare reported 5.1M trips in 2025 (5% growth); Citi Bike reported 38M (8% growth). Docked systems benefit from station infrastructure that creates predictability for riders and meters demand-side adoption.
+
+Dockless bike-share (without fixed stations) is largely consolidated; the experimentation phase ended in 2023. Lyft operates the dominant national network through its acquired bike-share division, with regional players in select markets. Operating economics for dockless are structurally weaker than docked due to vehicle redistribution costs, vandalism rates, and the absence of station-driven advertising revenue.
+
+A notable trend is the convergence of bike-share and dockless e-bike subscription models. Several operators now offer monthly memberships that include unlimited 30-minute trips on dockless e-bikes within a service zone. Adoption is concentrated in dense urban cores where car-free lifestyles are practical.
+
+---
+
+## Section 7: Charging Infrastructure Trends
+
+Charging infrastructure for e-bikes and e-scooters has emerged as a meaningful sub-segment, growing 28% in 2025. The dominant form factor remains residential at-home wall chargers (87% of installed base), but commercial charging — at workplaces, transit stations, and apartment buildings — is the fastest-growing sub-segment.
+
+Standardization remains a constraint. Battery interfaces have not converged; Bosch, Shimano, and various proprietary systems coexist. The European Union's USB-C mandate for portable electronics has not yet extended to e-mobility; industry observers expect regulatory pressure to follow within 3-5 years.
+
+Workplace charging is increasingly common in tech and creative-industry employers; we estimate 31% of large urban employers in tech-heavy metros now offer workplace e-bike charging, up from 12% in 2022. Apartment buildings lag — 7% of class-A multifamily properties offer common-area charging, with retrofit cost cited as the primary barrier.
+
+Public charging at transit hubs (subway/light rail stations) remains a stated priority across most major metro transit authorities, but actual installation lags policy commitments significantly. Funding fragmentation and permitting delays are the consistently cited bottlenecks.
+
+---
+
+## Section 8: Smart Helmet Category
+
+The connected safety hardware category — colloquially "smart helmets" — is the smallest segment we cover by retail value but has the strongest growth profile. The North American smart helmet market reached $810M in retail value in 2025, up from $480M in 2023, representing a 30% CAGR. We forecast $2.4B by 2030, contingent on the resolution of two open questions detailed below.
+
+**Category definition.** We define "smart helmets" as helmets that include at least one connected safety feature: turn signals (typically wireless-controlled), braking lights (auto-activated via accelerometer), crash detection (auto-notification to emergency contacts on detected impact), or integrated navigation/audio (bone-conduction speakers, often paired with smartphone apps). Helmets with passive integrated lighting only (no connectivity) are excluded from this category and tracked under traditional helmet retail.
+
+**Key players.** The category remains fragmented; no single manufacturer commands more than 15% market share. Top five by 2025 retail volume: Lumos Helmet (US, market leader at ~14% share with strong DTC presence), Sena Technologies (Korea, intercom heritage, ~11%), Coros (US/China, multi-sport, ~9%), Specialized ANGi (US, premium tier at ~7%), and POC Aid (Sweden, premium safety positioning at ~6%). Approximately 30 smaller brands hold the remaining share.
+
+**Crash detection technology.** Two architectures dominate: single-accelerometer crash detection (lower cost, higher false-positive rate) and multi-sensor fusion (accelerometer + gyroscope + GPS movement signature, lower false-positive rate but higher BOM cost). Insurance industry sources indicate that multi-sensor systems are likely to become a baseline requirement for any insurance discount programs, given that single-accelerometer systems triggered roughly 1 false alert per 47 hours of riding in our test panel.
+
+**Regulatory landscape.** Smart helmets sit at the intersection of two regulatory regimes: the Consumer Product Safety Commission's bicycle helmet standard (16 CFR 1203, governing impact protection) and the Federal Communications Commission's regulation of intentional radiators (governing the radio components for Bluetooth/cellular). Compliance with both is non-trivial. Eight smart helmet brands have had FCC Part 15 violations issued since 2023, typically for emissions exceeding limits during compliance testing. EU markets additionally require EN 1078 certification for the helmet shell; this is widely held but adds 3-5 months to a typical product development timeline.
+
+**Insurance industry interest.** Major auto insurers (State Farm, Progressive, Geico, Nationwide) are actively piloting telematics-style discount programs for cyclists who use connected safety helmets. The proposed structure mirrors auto-insurance "good driver" discount frameworks, with discounts of 5-15% on cycling-specific insurance riders or umbrella policies. As of Q1 2026, three insurers have public pilot programs and one (Progressive) has announced general availability for 2027. This could materially accelerate category adoption if discounts materialize at the upper end of the proposed range.
+
+**Distribution.** D2C dominates at 58% of retail value, reflecting the still-emerging category and the absence of strong channel inventory in independent bike shops. The specialty bike shop channel is growing rapidly (up from 12% to 22% of retail value over 2023-2025) as the category gains category-management attention from major distributors. Big-box channels (REI, Dick's Sporting Goods) are present but shallow in selection — typically 4-8 SKUs versus 40+ in dedicated specialty.
+
+**Open questions for the segment.** Our growth forecast is conditioned on (a) the proportion of insurers that follow Progressive into general availability of connected-safety discounts; (b) whether multi-sensor crash detection becomes a category baseline (lifting ASP) or remains a premium-tier feature; and (c) whether the current high false-positive rate of single-accelerometer systems triggers a consumer backlash that suppresses category trust before insurance discounts arrive. The downside scenario produces a 2030 category size of $1.4B versus our base-case $2.4B.
+
+---
+
+## Section 9: Vehicle-to-Grid Integration
+
+Vehicle-to-grid (V2G) integration of e-bike and e-scooter batteries is an emerging area, but practical commercial deployment is years away. The thesis is that fleet-scale dockless e-bikes and e-scooters represent meaningful aggregate battery capacity that could participate in demand-response markets, particularly in deregulated electricity markets.
+
+Several technical preconditions must be met: standardized battery interfaces (currently absent), bidirectional charging hardware (rare), aggregator software stack (early-stage), and regulatory clarity on energy market participation by mobility fleets (pre-policy). We treat this as a watch item for 2028+ rather than a current investable theme.
+
+---
+
+## Section 10: Outlook
+
+Our base-case forecast for North American e-mobility is $22.5B by 2030, with the e-bike segment reaching $11.8B (the largest), connected safety hardware reaching $2.4B (the fastest-growing in percentage terms), and charging infrastructure reaching $4.2B (driven by commercial and multifamily retrofit demand). Bike-share and dockless mobility plateau in the $2.5-3.0B range as urban density limits adoption ceilings.
+
+The largest single uncertainty in this forecast is the trajectory of insurance industry adoption of connected-safety telematics, which could accelerate or substantially constrain the smart helmet segment and, secondarily, influence rider behavior across the broader category. We will revisit forecasts in our Q4 2026 update.
+
+---
+
+*This report is prepared for the exclusive use of Meridian Insights subscribers. Reproduction or external distribution without written permission is prohibited.*

evals/bmm-skills/bmad-product-brief/files/mossridge-brief/addendum.md

@@ -0,0 +1,41 @@
+# Addendum — Mossridge Tool Lending Library
+
+## Options considered
+
+### Paid lending model (rejected)
+
+Considered charging a nominal per-loan fee ($2–$5) to cover replacement and maintenance. Rejected as inconsistent with library mission of free access. Board has previously stated free access is non-negotiable for core services. A donation jar at checkout was proposed as a soft alternative; deferred.
+
+### Hardware store partnership (considered, deferred)
+
+Mossridge Hardware (the store committing in-kind donations) offered to host a satellite lending point. Considered; deferred to year 2. The integration adds operational complexity (split inventory, cross-location tracking) we are not equipped for at launch. Reasonable to revisit once the main location is established.
+
+### Mobile lending van (rejected)
+
+Proposed by a board member to serve outlying areas. Rejected for MVP — capital cost ($35K+ for vehicle + outfitting) exceeds the entire grant. Could be a year-three expansion if demand validates.
+
+### Skills classes alongside tool loans (deferred)
+
+Considered offering "how to use a power drill" classes as a value-add. Deferred — interesting but distinct programming, not part of the lending service's MVP scope. Adult Services Librarian is interested in piloting separately.
+
+## Reference programs reviewed
+
+- Berkeley Tool Lending Library (operating since 1979, ~3,000 tools, 250+ daily loans). Funded as a city service.
+- Oakland Tool Lending Library (operating since 2000, smaller catalog, library-staffed).
+- Toronto Tool Library (nonprofit, member-supported, paid model — different funding architecture).
+
+Direct correspondence with Berkeley TLL staff (March 2026) suggested:
+- Theft has been low (~2% annually) due to library card requirement and community norms
+- The biggest sustainability risk has been staff hours, not tool replacement
+- Most successful programs have a paid coordinator role, not pure volunteer
+
+## Potential expansion (year 2+)
+
+- Hardware store satellite location
+- Specialty tool categories: woodworking, automotive, sewing
+- Skills classes paired with relevant tool checkouts
+- Seed/cuttings library co-located in spring/summer
+
+## Insurance and liability — current state
+
+Library counsel (Town of Mossridge legal department) has been consulted informally. Formal opinion pending. Existing policy covers patrons in the building; coverage for tool use off-premises is the open question. Awaiting written response before submitting grant application.

evals/bmm-skills/bmad-product-brief/files/mossridge-brief/brief.md

@@ -0,0 +1,57 @@
+---
+title: Mossridge Public Library — Tool Lending Library Proposal
+status: final
+created: 2026-04-30
+updated: 2026-04-30
+---
+
+# Tool Lending Library at Mossridge Public Library
+
+## What we're proposing
+
+A free tool-lending service operated out of the Mossridge Public Library, modeled on similar programs in Berkeley, Oakland, and Toronto. Cardholders borrow hand and power tools (drills, saws, ladders, sanders, plumbing snakes, gardening tools) for up to seven days, free of charge.
+
+## Why now
+
+Mossridge residents face rising costs of home maintenance and DIY supplies. Anecdotally, demand for community-shared resources is high — staff have fielded "do you lend tools?" requests for years. A tool library extends the library's mission of equitable access to information and skill-building into the practical-skills domain.
+
+## Who it serves
+
+Mossridge residents with active library cards. Primary audience: single-family homeowners doing their own home repairs, renters making minor improvements with landlord permission, hobbyist woodworkers and gardeners. Estimated 8,000 households in the library's service area.
+
+## Service design
+
+- **Catalog:** Approximately 200 tools to start, prioritizing the most-requested categories (drilling, cutting, sanding, ladders, garden).
+- **Loan period:** Seven days, one renewal allowed if no holds.
+- **Borrower requirements:** Active library card, signed liability waiver, completed safety briefing for power tools.
+- **Location:** Library basement, currently underutilized storage. Accessible by elevator.
+- **Hours:** Tuesday–Saturday during library hours; tools returned via after-hours drop slot when closed.
+
+## Funding
+
+- ARPA infrastructure grant: $42,000 (anticipated, application pending)
+- Friends of the Mossridge Library matching funds: $10,000 (committed)
+- In-kind tool donations from Mossridge Hardware (committed in principle)
+
+Year-one operating cost is estimated at $48,000, primarily tool purchase, maintenance supplies, and shelving/storage retrofit. Ongoing cost (year two and beyond) projected at $12,000 annually for replacement tools and consumables.
+
+## Operations
+
+The service will be run by trained library volunteers, supervised by the Adult Services Librarian. Volunteer training program to be developed in partnership with Mossridge Vocational Center. Estimated 4–6 active volunteers needed at any given time, with a roster of 12–15 trained volunteers to provide coverage.
+
+## Risks
+
+- **Theft and loss.** Tools are valuable and portable. Mitigation: deposit on power tools (refundable), card-required checkout, photo documentation at loan and return.
+- **Liability.** Borrower waivers will be required; the library's existing insurance policy is being reviewed for coverage.
+- **Demand uncertainty.** We do not yet know the actual borrowing volume the service will see.
+
+## Success criteria
+
+- Launch by Q3 2027 with a catalog of 200 tools.
+- 300 unique borrowers in the first year of operation.
+- Zero serious injury incidents.
+- Tool loss rate under 5% per year.
+
+## What we're asking
+
+Board approval to proceed with the ARPA grant application and finalize the service design for fall 2027 launch.

evals/bmm-skills/bmad-product-brief/files/mossridge-brief/decision-log.md

@@ -0,0 +1,29 @@
+# Decision Log — Mossridge Tool Lending Library
+
+## 2026-03-04
+- **Pursuing the project.** Adult Services Librarian + Library Director agreed there's enough informal demand signal (years of "do you lend tools?" inquiries) to investigate seriously. Acknowledged that informal inquiries are not the same as validated demand.
+
+## 2026-03-11
+- **Reference programs to study: Berkeley, Oakland, Toronto.** Selected based on size, longevity, and accessibility of operational data.
+
+## 2026-03-25
+- **Initial scope: hand and power tools only.** Rejected including specialty categories (sewing, electronics test gear, automotive) for MVP. Reason: staff expertise and storage. Revisit year 2.
+- **Free model.** Confirmed — paid model rejected as inconsistent with library mission. Donation jar approved as soft revenue.
+
+## 2026-04-01
+- **Volunteer-run model.** Selected to keep ongoing operating costs low. Acknowledged risk: Berkeley correspondence flagged staff-hours as the biggest sustainability concern in similar programs. Plan to revisit at year-one review.
+
+## 2026-04-08
+- **Funding architecture: ARPA grant + Friends matching + in-kind donations.** Considered municipal budget request; rejected as too slow (next budget cycle is 18 months out). Grant is faster but requires fall 2027 launch deadline.
+
+## 2026-04-15
+- **Launch timing: Q3 2027.** Driven by ARPA grant deadline, not by service-readiness analysis. Acknowledged this is grant-driven, not user-driven, timing.
+- **Year-one target: 300 unique borrowers.** Set by analogy to comparable programs scaled to Mossridge population. No local validation underlying this number.
+
+## 2026-04-22
+- **Hardware store satellite deferred to year 2.** Operational complexity exceeds our launch capacity.
+- **Liability: pending formal opinion from town legal.** Borrower waiver in draft.
+
+## 2026-04-30
+- **Brief finalized for board meeting.** Status moved to final.
+- **Open items acknowledged for board discussion:** demand validation method, volunteer sustainability, written legal opinion on off-premises tool use coverage.

evals/bmm-skills/bmad-product-brief/files/pantry-bridge-interviews.md

@@ -0,0 +1,90 @@
+# Pantry Bridge — Customer Research Transcripts
+
+**Project:** Pantry Bridge meal-kit concept exploration
+**Research firm:** In-house
+**Round:** Discovery interviews, March 2026
+**Format:** 45-minute semi-structured interviews, video; excerpts below are lightly edited for length and clarity
+
+The four interviews below cover four distinct potential customer segments. We are sharing all four for context, though the team's current product hypothesis targets one specific segment.
+
+---
+
+## Interview 1 — Susan, 38, working parent
+
+**Household:** Two kids (ages 6 and 9), spouse works full-time, both parents work demanding office jobs. Suburban Chicago.
+
+**Susan:** "Honestly, the question is just — can I get dinner on the table by 6:30 without it being chicken nuggets again? My kids don't eat anything green unless we play games about it. My husband and I both have late meetings sometimes. We've tried HelloFresh, we've tried Blue Apron, we tried Home Chef. They all kind of work, and they all kind of don't.
+
+The thing that breaks them for us is the prep time. The boxes say 30 minutes but you need to add 10-15 to actually get it done. By Wednesday night I don't have 45 minutes. So we end up using the boxes on weekends and ordering takeout three nights a week, which is the opposite of what the boxes are supposed to do.
+
+If you really wanted to crack it for families like ours: pre-chopped vegetables, sauces that are actually finished and not 'whisk these eight things together.' I'll pay more for less prep. And the recipe books need to read like the kid is going to eat it — not like 'spicy harissa-rubbed cauliflower steaks.'
+
+Portion sizing — most kits send way too much for our family. We're a family of four but the kids each eat about 60% of a meal. We end up with leftovers that go bad. Better sizing would help."
+
+**Interviewer:** What about price?
+
+**Susan:** "We spend $250-350 a week on groceries currently and probably another $200 on takeout. So a meal kit that replaces three nights of takeout could be $200 a month and we'd still come out ahead. Most kits are priced fine; it's the time that breaks them."
+
+---
+
+## Interview 2 — Marcus, 21, college student
+
+**Household:** Junior at state university, off-campus apartment shared with two roommates, kitchen has a microwave, a stovetop, and a half-broken oven. Limited budget.
+
+**Marcus:** "I'm probably the wrong person for this conversation, no offense. I'm not really a meal-kit person. My food situation is, like, dining hall meal plan when I can use it, and the rest is whatever's cheap and fast. Trader Joe's frozen stuff. Eggs. Pasta. Costco runs with my roommates once a month.
+
+I tried a meal kit when my mom signed me up as a 'starting college' gift. It was nice, but it was $80 a week for two people, which is way out of budget. And honestly, the thing they don't get is that I don't have time at 7 PM to cook. I have time at 11 PM. I want to grab something on my way back from the library and not think.
+
+If you're trying to do meal kits for college students — and I don't really think you should — but if you were, the price has to be like $5 a meal. And it has to be food that survives in a fridge for two weeks because we don't shop on a weekly schedule. We shop when we run out.
+
+Snacks matter more to us than meals, actually. Like, the moment when I'm desperate is 10 PM in the library, not 7 PM. Solve that and I might pay attention."
+
+**Interviewer:** Do you have any dietary restrictions?
+
+**Marcus:** "I'm vegetarian, sort of. I eat fish. So pescatarian I guess. But mostly because meat is expensive."
+
+---
+
+## Interview 3 — Eleanor, 71, retired, lives alone
+
+**Household:** Widow, lives alone in the same single-family home she's been in for 36 years. Suburban Cleveland. Two adult children live out of state. Drives during the day but no longer at night.
+
+**Eleanor:** "I'll tell you what I miss. I miss cooking for someone. My husband Walter passed five years ago this June, and the hardest thing — well, not the hardest, but one of them — is that I don't really cook anymore. I cook eggs. I cook a piece of fish. I open a can of soup more often than I'd like to admit. I used to make Sunday dinners that would feed eight people. Now I eat standing up at the counter half the time.
+
+The grocery store is genuinely difficult. I drive there, I park in the back of the lot because I can usually find a spot, and then it's a long walk in. I get tired by the time I'm in the dairy aisle. Carrying the bags from the car to the kitchen — that's a project. My daughter wants me to use grocery delivery and I've tried, but the apps are all designed for someone twenty years younger than me. Tiny buttons, asking me to click through six screens to add a single tomato. I get frustrated and give up.
+
+What I would actually want — and I've thought about this — is meals for one person. Real portions. Not a frozen TV dinner. Not 'serves four, freeze the rest.' I have a freezer full of leftovers I'll never eat. Just one good meal that I can heat up or finish cooking, that tastes like food I would have made.
+
+I'm watching my sodium because of my blood pressure. Watching sugar too — borderline diabetic, my doctor calls it. So I read labels carefully. The frozen meals you can buy in stores are loaded with both. I'd pay more for less of both, if I trusted that the labels were accurate.
+
+The other thing — and please put this in your notes — is that I'm careful about who I let into my house and what I sign up for. There are scams. My friend Marian got taken for $4,000 last year. So if some company asks for my information, I want to know who they are. I want a real customer service number with a real person. I want it to feel like a real business, not a flashy app.
+
+I don't want it to feel like 'old-people food.' That's an important thing. The Meals on Wheels program in our township is wonderful but it's clearly designed for people who are sicker than I am. I'm not sick. I just live alone and grocery shopping is a lot."
+
+**Interviewer:** What would the ideal experience look like?
+
+**Eleanor:** "Someone delivers good food, in real portions, made with the kind of ingredients I would have used. I can heat it up or finish it. It doesn't taste like a hospital. The packaging is something I can actually open without a knife. I get a phone call once in a while from a person, not a robot. The price is reasonable — I'm on a fixed income but I can spend on things that matter. Eating well matters."
+
+---
+
+## Interview 4 — Dimitri, 44, Director of Food Services, mid-size hospital
+
+**Organization:** 340-bed hospital, food service operates patient meals, staff cafeteria, and a small retail café. Reports to the COO.
+
+**Dimitri:** "I'm probably also not who you should be talking to, but happy to share. We don't buy meal kits. We buy ingredients in institutional volumes from Sysco and US Foods primarily, with some specialty buys for dietary restrictions. We feed about 1,800 people a day across patients, staff, and visitors.
+
+What I deal with that you might find interesting is the patient diet matrix. We have to produce meals that meet specific medical requirements — renal diets, cardiac diets, diabetic diets, dysphagia textures, allergen-free, religious restrictions. Each patient gets a tray that meets their specific orders. It's complex.
+
+If a meal kit company wanted to play in our world, they'd be selling to me at the institutional level — bulk pricing, multi-year contracts, ability to deliver consistent specs across thousands of meals. That's not really a 'meal kit' anymore; that's wholesale food service.
+
+Now, where I might be a buyer in a different sense: my staff cafeteria. We're trying to compete with grab-and-go culture. If you produced ready-to-heat meals targeting our staff demographic — nurses, doctors, techs, who are working 12-hour shifts and want real food, not a sandwich — I might pay attention. But the price point would have to make sense for institutional buying, and you'd need to integrate with our existing food safety protocols.
+
+For consumer meal kits, I'm probably not your customer. We did try one when my wife and I were both working through COVID, and we let the subscription lapse after about three months. Fine product, just didn't fit our patterns."
+
+---
+
+## Note from the research lead
+
+These four interviews were selected to represent the range of segments we've considered. The team's working hypothesis after this round is that the older-adult-living-alone segment is the strongest fit for the Pantry Bridge concept — distinctive needs, acknowledged friction with current options, willingness to pay for quality, and a meaningful unmet need around portion sizing and trust. Working parent segment is well-served by existing competitors. College student segment is too price-sensitive. Institutional segment is a different business entirely.
+
+The brief should target the older-adult segment based on the Eleanor interview specifically.

evals/bmm-skills/bmad-product-brief/files/q2-brainstorm.md

@@ -0,0 +1,101 @@
+# Q2 Brainstorm — Hatchet & Loop Studio
+
+**Date:** 2026-04-15
+**Present:** Mira, Devon, Sofia, Theo
+
+Annual Q2 ideation. We're hunting for our next side-project-that-could-become-a-product. Format: 10 minutes wild ideas, 3 minutes per idea on quick takes, then we vote on one to dig into.
+
+## Round 1: Everything goes
+
+(10 minutes, no filtering. We just throw stuff out.)
+
+- A weather app that tracks your mood alongside the forecast (Devon)
+- Meditation chime that learns your sleep cycle and chimes only at the right wake-window (Theo)
+- A podcasting tool for non-podcasters — like, you record voice notes and it auto-edits and posts (Sofia)
+- Craft beer subscription with detailed brewer notes you can read while drinking (Mira)
+- AI sommelier app that tells you what wine to buy at Trader Joe's based on a photo (Theo)
+- Office-plant-care subscription with auto-replacement when one dies (Devon)
+- Neighborhood ride coordinator — like a private Uber pool for one neighborhood (Mira)
+- Neighborhood compost coordinator — connect people with food scraps to people with active compost piles (Sofia)
+- Cookbook app where you click "I'll cook this Tuesday" and it auto-generates the shopping list and sends it to your delivery service (Devon)
+- AR home staging — point your phone at a room and it shows you what it would look like with different furniture (Theo)
+
+## Round 2: Quick takes
+
+### Weather + mood
+
+Devon: "I'd use it." Sofia thinks the data correlation isn't strong enough to be useful — interesting concept but the science doesn't support a product. Park.
+
+### Sleep-cycle meditation chime
+
+Theo's pitch — exists already (Sleep Cycle, etc.). Differentiation would be the chime, which is hardware. Out of scope for a software-first studio.
+
+### Podcasting for non-podcasters
+
+Sofia: "There are like fifty of these." She's right. Skip.
+
+### Craft beer subscription
+
+Mira admits this is mostly her wanting it for herself. We're not in the logistics business. Skip.
+
+### AI sommelier
+
+Theo: "The model would have to be incredibly good at label recognition." Sofia: "And there's already Vivino." Skip.
+
+### Office-plant-care subscription
+
+Devon: "I worked at a place that had this. They were always sad plants." Operational nightmare, low margin. Skip.
+
+### Neighborhood ride coordinator
+
+Mira: "Saturated. Lyft and Uber both have pool features. Uber Neighborhood was a thing and they killed it." Skip.
+
+### Neighborhood compost coordinator
+
+Sofia: "Hear me out. Cities are mandating organic waste separation but most apartments don't have a composting option. People in single-family homes often have active compost piles and would love more material. There's a missing match-making layer." General agreement this is more interesting than the others. Theo: "How do we make money?" Sofia: "Eventually a small fee on the compost-pile-host side, but for MVP just free and prove the demand." Group lights up. We agree to dig into this in Round 3.
+
+### Cookbook → shopping list
+
+Devon's pitch. Already exists (Mealime, Plan to Eat). Skip.
+
+### AR home staging
+
+Theo: "IKEA already has this." Skip.
+
+## Round 3: Compost coordinator deep dive
+
+We spent 45 minutes on this. Notes:
+
+**Who is the user?**
+Two-sided market. Side A: apartment dwellers and renters who generate food scraps and want them composted (motivated by environmental values, sometimes by city mandates). Side B: people with active backyard compost piles who want more "browns and greens" — single-family homeowners, urban farmers, school gardens, community gardens.
+
+Sofia thinks Side A is the harder side to acquire (weak intent — recycling-adjacent behavior). Side B is easier but smaller. The product has to be designed around Side A's friction points.
+
+**Geographic scope.**
+Hyperlocal — neighborhood-level, not city-wide. The whole point is short-distance handoff: Side A doesn't want to drive their food scraps across town. We're talking 5-block radius matches.
+
+**Business model (later).**
+Free at launch. Eventually: subscription for Side B (compost-pile hosts) — they pay to access more matches. Side A always free. Possibly partner with cities that have green-waste mandates (B2G channel).
+
+**Technical approach.**
+Web app first, mobile second. Map-based discovery. Identity verification light-touch (apartment dwellers are skittish about strangers; need trust signals). Match-and-message pattern, not real-time logistics.
+
+**Competition.**
+ShareWaste exists but is global and not focused on hyperlocal density. Some city-specific apps (NYC's GrowNYC). No one has cracked the neighborhood-density model.
+
+**MVP scope.**
+One pilot neighborhood. Sofia knows people in a Portland neighborhood (Sunnyside / Hawthorne area) where compost culture is strong. Start there.
+
+**Open questions.**
+- How do we acquire Side A (apartment dwellers)? They have low intent and lots of competing options (just throwing scraps in trash, paying a service, signing up for city pickup if available).
+- What does the trust layer look like? Reviews? Vouching? Real-name only?
+- Does Side B saturation become a problem fast (one compost pile can only take so much)? How do we route demand?
+
+## Action items
+
+- Sofia: write up the compost coordinator concept as a brief by next Wednesday. Take it to Mira and Devon for first read.
+- Devon: research ShareWaste's user numbers and any teardowns of why they haven't dominated.
+- Theo: sketch the trust-layer UX concepts.
+- Mira: talk to Sofia's Portland contacts about doing user interviews.
+
+Next meeting: 2026-04-29 — review brief draft, decide on go/no-go.

evals/bmm-skills/bmad-product-brief/triggers.json

@@ -0,0 +1,18 @@
+[
+  { "query": "Help me write a product brief for my new app idea", "should_trigger": true },
+  { "query": "I need to draft a brief for a feature we're scoping", "should_trigger": true },
+  { "query": "Update this product brief — we changed the target audience", "should_trigger": true },
+  { "query": "Review my brief and tell me if it's investor-ready", "should_trigger": true },
+  { "query": "Validate this brief before our board meeting Monday", "should_trigger": true },
+  { "query": "Pressure-test my product brief for weak assumptions", "should_trigger": true },
+  { "query": "Help me put together a one-page summary of my product idea for stakeholders", "should_trigger": true },
+
+  { "query": "Help me brainstorm ideas for a new feature", "should_trigger": false },
+  { "query": "Write me a PRD for our checkout flow redesign", "should_trigger": false },
+  { "query": "Run a working backwards exercise for my product idea", "should_trigger": false },
+  { "query": "Document this existing codebase for AI agents", "should_trigger": false },
+  { "query": "Help me write user stories for the next sprint", "should_trigger": false },
+  { "query": "Generate a system architecture for my app", "should_trigger": false },
+  { "query": "Write code to parse JSON in Python", "should_trigger": false },
+  { "query": "Create a marketing landing page for my product", "should_trigger": false }
+]

package.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://json.schemastore.org/package.json",
   "name": "bmad-method",
-  "version": "6.9.0",
+  "version": "6.8.0",
   "description": "Breakthrough Method of Agile AI-driven Development",
   "keywords": [
     "agile",
@@ -67,12 +67,6 @@
       "markdownlint-cli2"
     ]
   },
-  "overrides": {
-    "esbuild": "^0.28.1",
-    "markdownlint-cli2": {
-      "markdown-it": "^14.2.0"
-    }
-  },
   "dependencies": {
     "@clack/core": "^1.3.1",
     "@clack/prompts": "^1.4.0",
@@ -89,10 +83,10 @@
     "yaml": "^2.7.0"
   },
   "devDependencies": {
-    "@astrojs/sitemap": "^3.7.3",
-    "@astrojs/starlight": "^0.40.0",
+    "@astrojs/sitemap": "^3.6.0",
+    "@astrojs/starlight": "^0.37.5",
     "@eslint/js": "^9.33.0",
-    "astro": "^6.4.6",
+    "astro": "^5.16.0",
     "c8": "^10.1.3",
     "eslint": "^9.33.0",
     "eslint-config-prettier": "^10.1.8",

src/bmm-skills/1-analysis/bmad-prfaq/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: bmad-prfaq
-description: Working Backwards PRFAQ challenge that stress-tests a product concept customer-first. Use when the user requests to 'create a PRFAQ', 'work backwards', or 'run the PRFAQ challenge'.
+description: Working Backwards PRFAQ challenge to forge product concepts. Use when the user requests to 'create a PRFAQ', 'work backwards', or 'run the PRFAQ challenge'.
 ---
 
 # Working Backwards: The PRFAQ Challenge
@@ -107,7 +107,7 @@ When the user gets stuck, offer concrete suggestions based on what they've share
 
 **Fast-track:** If the user provides all four essentials in their opening message (or via structured input), acknowledge and confirm understanding, then move directly to document creation and Stage 2 without extended discovery.
 
-**Graceful redirect:** If after 2-3 exchanges the user can't articulate a customer or problem, don't force it. Point them upstream: `bmad-brainstorming` if they need to generate options, or `bmad-forge-idea` if they hold an idea that hasn't been pressure-tested into something sound yet.
+**Graceful redirect:** If after 2-3 exchanges the user can't articulate a customer or problem, don't force it — suggest the idea may need more exploration first and recommend they invoke the `bmad-brainstorming` skill to develop it further.
 
 **Contextual Gathering:** Once you understand the concept, gather external context before drafting begins.
 

src/bmm-skills/1-analysis/bmad-product-brief/SKILL.md

@@ -15,7 +15,7 @@ At the opening greeting, let the user know they can invoke `bmad-party-mode` for
 
 ## On Activation
 
-1. Resolve customization: `uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly and use defaults.
+1. Resolve customization: `python3 {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly and use defaults.
 2. Execute each entry in `{workflow.activation_steps_prepend}` in order.
 3. Treat every entry in `{workflow.persistent_facts}` as foundational context for the rest of the run. Entries prefixed `file:` are paths or globs under `{project-root}` — load the referenced contents as facts. All other entries are facts verbatim.
 4. `{workflow.external_sources}` is an org-configured registry of internal tools (knowledge bases, MCP tools); consult them alongside generic web research on the same triggers in `## Discovery`, org tools preferred when their directive matches. If a named tool is unavailable at runtime, fall back to standard behavior and note the gap when relevant.
@@ -28,11 +28,11 @@ Activation is complete. If `activation_steps_prepend` or `activation_steps_appen
 
 ## Intent Operating Modes
 
-**Create.** A brief the user is proud of, that meets their needs, drawn out through real conversation — do not assume: instead converse and understand, and then help craft the best product brief for their needs. Begin in `## Discovery` before drafting; the brief comes after the picture is on the table. Shape follows the product and need. Treat `{workflow.brief_template}` as a starting structure, not a contract: drop sections that do not earn their place, add sections the product needs, reorder freely - create sections for specialized domains or concerns also as needed. The brief serves the product's story, not the template's shape. Bind `{doc_workspace}` to a fresh folder at `{workflow.brief_output_path}/{workflow.run_folder_pattern}/`, write `brief.md` there with YAML frontmatter (title, status, created, updated), and seed the memlog: `uv run {project-root}/_bmad/scripts/memlog.py init --workspace {doc_workspace} --field topic="<product>"`. For Update and Validate, `{doc_workspace}` is the existing folder of the brief being targeted.
+**Create.** A brief the user is proud of, that meets their needs, drawn out through real conversation — do not assume: instead converse and understand, and then help craft the best product brief for their needs. Begin in `## Discovery` before drafting; the brief comes after the picture is on the table. Shape follows the product and need. Treat `{workflow.brief_template}` as a starting structure, not a contract: drop sections that do not earn their place, add sections the product needs, reorder freely - create sections for specialized domains or concerns also as needed. The brief serves the product's story, not the template's shape. Bind `{doc_workspace}` to a fresh folder at `{workflow.brief_output_path}/{workflow.run_folder_pattern}/` and write `brief.md` there with YAML frontmatter (title, status, created, updated). For Update and Validate, `{doc_workspace}` is the existing folder of the brief being targeted.
 
-**Update.** Reconcile an existing brief with a change signal. Before proposing changes, read the brief, addendum, `.memlog.md`, and original inputs — and run the `## Discovery` posture against the change signal (a patch applied without context becomes drift). If `.memlog.md` is missing (a legacy or pre-standard brief), init it with `uv run {project-root}/_bmad/scripts/memlog.py init --workspace {doc_workspace}` first — this update is its first entry. Surface conflicts with prior decisions before changing. Headless override: log the reversal via `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type override --text "<reversal + rationale>"`, then apply; halt `blocked` if intent is ambiguous. If the change is fundamental, offer Create instead of patching.
+**Update.** Reconcile an existing brief with a change signal. Before proposing changes, read the brief, addendum, `.decision-log.md`, and original inputs — and run the `## Discovery` posture against the change signal (a patch applied without context becomes drift). Surface conflicts with prior decisions before changing. Headless override: log the reversal to `.decision-log.md`, then apply; halt `blocked` if intent is ambiguous. If the change is fundamental, offer Create instead of patching.
 
-**Validate.** Honest critique against the brief's own purpose. Read the brief, the addendum if present, `.memlog.md`, and any original inputs first — a validation that ignores prior decisions, rejected ideas, or context the user supplied is shallow. Cite specific lines. Caveat what cannot be evaluated. Return inline — no separate file unless asked. Always offer to roll findings into an Update, even in headless mode — include `"offer_to_update": true` in the JSON status block.
+**Validate.** Honest critique against the brief's own purpose. Read the brief, the addendum if present, `.decision-log.md`, and any original inputs first — a validation that ignores prior decisions, rejected ideas, or context the user supplied is shallow. Cite specific lines. Caveat what cannot be evaluated. Return inline — no separate file unless asked. Always offer to roll findings into an Update, even in headless mode — include `"offer_to_update": true` in the JSON status block.
 
 ## Headless Mode
 
@@ -44,7 +44,7 @@ When invoked headless, do not ask. Complete the intent using what is provided, w
   "intent": "create",
   "brief": "{doc_workspace}/brief.md",
   "addendum": "{doc_workspace}/addendum.md",
-  "memlog": "{doc_workspace}/.memlog.md",
+  "decision_log": "{doc_workspace}/.decision-log.md",
   "open_questions": [],
   "external_handoffs": [
     {"directive": "Confluence upload", "tool": "corp:confluence_upload", "url": "https://confluence.corp/PROD/123", "status": "ok"}
@@ -76,15 +76,15 @@ The workspace persists; stop and resume freely. The opener's philosophy (not in
 ## Constraints
 
 - **Right-size to purpose.** A passion project does not need investor-grade rigor. A VC pitch input does. Read the room.
-- **Persistence is real-time.** Once Create intent is confirmed, the workspace (run folder, `brief.md` skeleton with `status: draft`, `.memlog.md` seeded via `memlog.py init`) exists on disk and the user knows the path.
-- **File roles.** `.memlog.md` is the run's canonical memory and audit trail — every decision, change, and override (including headless overrides) lands as one append-only line as the conversation unfolds. All writes go through the shared script, never by hand: `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type <decision|change|override|assumption|event> --text "<one-line gist, reason included>"` (atomic; read it back only to resume or audit). The brief is distilled toward it; whatever isn't logged is lost on resume. `addendum.md` preserves user-contributed depth that belongs in a downstream document (PRD, architecture, solution design) or earned a place but does not fit the brief (rejected-alternative rationale, options-considered matrices, parked-roadmap context, technical constraints, in-depth personas, sizing data). Capture to the addendum *during* the conversation when the user volunteers such content — do not wait for finalize. Audit and override information never goes in the addendum.
+- **Persistence is real-time.** Once Create intent is confirmed, the workspace (run folder, `brief.md` skeleton with `status: draft`, `.decision-log.md`) exists on disk and the user knows the path.
+- **File roles.** `.decision-log.md` is canonical memory and audit trail — every decision, change, and override (including headless overrides) is recorded there as the conversation unfolds. `addendum.md` preserves user-contributed depth that belongs in a downstream document (PRD, architecture, solution design) or earned a place but does not fit the brief (rejected-alternative rationale, options-considered matrices, parked-roadmap context, technical constraints, in-depth personas, sizing data). Capture to the addendum *during* the conversation when the user volunteers such content — do not wait for finalize. Audit and override information never goes in the addendum.
 - **Continuity across sessions.** If a prior in-progress draft for this project exists, the user is offered to resume.
 - **Extract, don't ingest.** Source artifacts (provided by the user or discovered during the run — transcripts, brainstorms, research reports, code, web results, prior briefs) enter the parent conversation as relevance-filtered extracts, not loaded wholesale. Subagents do the extraction against the user's stated focus; the parent context stays lean.
 - **Length and coherence.** Aim for 1-2 pages — if it is longer, the detail belongs in the addendum. Structure in service of the product; downstream consumers (PRD workflow, etc.) read this, so coherent shape matters.
 
 ## Finalize
 
-1. Memlog audit + addendum review: the user ends this step with an explicit, shared accounting of how the meaningful contents of `.memlog.md` were handled — captured in the brief, captured in `addendum.md` (which may already hold detail captured during the conversation — see `## Constraints` for what belongs there), or set aside as process noise.
+1. Decision log audit + addendum review: the user ends this step with an explicit, shared accounting of how the meaningful contents of `.decision-log.md` were handled — captured in the brief, captured in `addendum.md` (which may already hold detail captured during the conversation — see `## Constraints` for what belongs there), or set aside as process noise.
 2. Polish: apply each entry in `{workflow.doc_standards}` (a `skill:`, `file:`, or plain-text directive) to `brief.md` (and `addendum.md` if it exists). Run passes as parallel subagents - apply all doc standards to `brief.md` first, then `addendum.md` so we present a high-quality draft for the user to review and finalize.
 3. External handoffs: execute each entry in `{workflow.external_handoffs}` to route artifacts beyond local files (Confluence, Notion, ticket systems, etc.) — each directive names the MCP tool and the fields it needs. Invoke the tool, capture any URLs or IDs returned, and surface them in the user message. If a named tool is unavailable, skip that handoff and flag it; local files always exist regardless.
 4. Tell the user it is ready: local paths and external destinations (URLs returned from handoffs). Invoke `bmad-help` to suggest what next steps make sense in the bmad method ecosystem.

src/bmm-skills/2-plan-workflows/bmad-prd/SKILL.md

@@ -11,11 +11,11 @@ You are a master facilitator and coach helping the user create, edit, or validat
 - Bare paths resolve from skill root; `{skill-root}` is this skill's install dir; `{project-root}` is the project working dir.
 - `{workflow.<name>}` resolves to fields in `customize.toml`'s `[workflow]` table (overrides win per BMad merge rules).
 - `{doc_workspace}` is the bound run folder.
-- **File roles.** `.memlog.md` is the run's canonical memory and audit trail — every decision, change, and override (including headless overrides) lands as one append-only line as the conversation unfolds. All writes go through the shared script, never by hand: `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type <decision|change|override|assumption|event> --text "<one-line gist, reason included>"` (atomic; read it back only to resume or audit). The PRD is distilled toward it; whatever isn't logged is lost on resume. `addendum.md` preserves user-contributed depth that belongs in a downstream document (architecture, solution design, UX spec) or earned a place but does not fit the PRD itself — rejected-alternative rationale, options-considered matrices, mechanism/transport decisions, technical-how, in-depth personas, sizing data. Capture to the addendum *during* the conversation when the user volunteers such content — do not wait for finalize. Audit and override information never goes in the addendum.
+- **File roles.** `.decision-log.md` is canonical memory and audit trail — every decision, change, and override (including headless overrides) is recorded there as the conversation unfolds. `addendum.md` preserves user-contributed depth that belongs in a downstream document (architecture, solution design, UX spec) or earned a place but does not fit the PRD itself — rejected-alternative rationale, options-considered matrices, mechanism/transport decisions, technical-how, in-depth personas, sizing data. Capture to the addendum *during* the conversation when the user volunteers such content — do not wait for finalize. Audit and override information never goes in the addendum.
 
 ## On Activation
 
-1. Resolve customization: `uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly and use defaults.
+1. Resolve customization: `python3 {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly and use defaults.
 2. Run `{workflow.activation_steps_prepend}`. Treat `{workflow.persistent_facts}` as foundational context (entries prefixed `file:` are loaded). `{workflow.external_sources}` is an org-configured registry of internal tools (knowledge bases, MCP tools); consult them alongside generic web research on the same triggers, org tools preferred when their directive matches. Research itself fires during Discovery — see **Research subagents**.
 3. Load `{project-root}/_bmad/bmm/config.yaml` (+ `config.user.yaml` if present). Resolve `{user_name}`, `{communication_language}`, `{document_output_language}`, `{planning_artifacts}`, `{project_name}`, `{date}`. Missing keys → neutral defaults; never block.
 4. If headless, follow `references/headless.md` for the whole run. Otherwise greet the user **by name** using `{user_name}` and **in their language** using `{communication_language}` — and stay in `{communication_language}` for every turn for the entire run, not just the greeting. In the greeting, let the user know that at any point they can invoke `bmad-party-mode` for multi-agent perspectives or `bmad-advanced-elicitation` for deeper exploration on a specific section. Then scan for misroute on the first message: if the signal points elsewhere (game → BMad GDS; express build → `bmad-quick-dev`; one-pager → `bmad-product-brief`; vet product idea → `bmad-prfaq`; agent skill or custom agent → `bmad-workflow-builder`), suggest they might want the other options before continuing.
@@ -27,9 +27,9 @@ Activation is complete. If `activation_steps_prepend` or `activation_steps_appen
 
 ## Intent Modes
 
-**Create.** Bind `{doc_workspace}` to `{workflow.prd_output_path}/{workflow.run_folder_pattern}/`. Write `prd.md` with YAML frontmatter (title, status, created, updated — initial `status: draft`), and seed the memlog with `uv run {project-root}/_bmad/scripts/memlog.py init --workspace {doc_workspace} --field topic="<PRD/product name>"` so subsequent decisions land in a known file. Tell the user the path. Run `## Discovery`, then `## Finalize`.
+**Create.** Bind `{doc_workspace}` to `{workflow.prd_output_path}/{workflow.run_folder_pattern}/`. Write `prd.md` with YAML frontmatter (title, status, created, updated — initial `status: draft`), and create the `.decision-log.md` skeleton at the workspace root so subsequent decisions land in a known file. Tell the user the path. Run `## Discovery`, then `## Finalize`.
 
-**Update.** Reconcile the PRD with a change signal. Source-extract against PRD, addendum, `.memlog.md`, and original inputs (extract, don't ingest). If `.memlog.md` is missing, init it with `uv run {project-root}/_bmad/scripts/memlog.py init --workspace {doc_workspace}`, then spawn a one-time bootstrap subagent to reverse-engineer a thin log from the PRD (one `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type decision --text "<recovered decision>"` per recovered decision) before continuing. Surface conflicts with prior decisions before applying. Then `## Finalize`.
+**Update.** Reconcile the PRD with a change signal. Source-extract against PRD, addendum, `.decision-log.md`, and original inputs (extract, don't ingest). If `.decision-log.md` is missing, spawn a one-time bootstrap subagent to reverse-engineer a thin log from the PRD before continuing. Surface conflicts with prior decisions before applying. Then `## Finalize`.
 
 **Validate** (or *analyze*). Critique without changing. Load `references/validate.md`.
 
@@ -82,11 +82,11 @@ Under Validate intent, the parent additionally runs the synthesis pipeline in `r
 
 Tell the user the sequence in one sentence, then walk it. Polish goes last so it does not redo work after reviewer fixes.
 
-1. **Memlog audit.** Walk `.memlog.md` with the user; each entry captured in PRD, in addendum, or set aside.
+1. **Decision log audit.** Walk `.decision-log.md` with the user; each entry captured in PRD, in addendum, or set aside.
 2. **Input reconciliation.** Subagent per user-supplied input against `prd.md` + `addendum.md`. Each writes its extract to `{doc_workspace}/reconcile-{slug}.md` and returns ONLY a compact summary (input name, gaps 2-5, file path). Surface gaps — especially qualitative ideas (tone, voice, feel) the FR structure silently drops. Must happen before polish.
 3. **Reviewer pass.** Run `## Reviewer Gate`. Resolve before polish.
-4. **Triage open items.** All Open Questions, `[ASSUMPTION]` tags, `[NOTE FOR PM]` callouts. Phase-blockers (would make the PRD unsafe for UX/architecture/epics) surfaced one at a time and resolved; non-blockers deferred with owner + revisit condition logged via `memlog.py append`. If phase-blocker count is high, flag it.
+4. **Triage open items.** All Open Questions, `[ASSUMPTION]` tags, `[NOTE FOR PM]` callouts. Phase-blockers (would make the PRD unsafe for UX/architecture/epics) surfaced one at a time and resolved; non-blockers deferred with owner + revisit condition logged to `.decision-log.md`. If phase-blocker count is high, flag it.
 5. **Polish.** Apply `{workflow.doc_standards}` to `prd.md` and `addendum.md` in declared order (structural passes before prose — prose should not polish soon-to-be-cut text). Parallelize across documents, sequential within.
 6. **External handoffs.** Execute `{workflow.external_handoffs}`; surface returned URLs/IDs. Skip and flag unavailable tools.
-7. **Close.** Set `prd.md` frontmatter `status: final` and `updated` to `{date}` so future invocations distinguish this PRD from in-progress drafts. Record finalization via `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type event --text "PRD finalized"`. Share artifact paths. Common next: `bmad-ux`, `bmad-architecture`, `bmad-create-epics-and-stories`; invoke `bmad-help` for authoritative routing.
+7. **Close.** Set `prd.md` frontmatter `status: final` and `updated` to `{date}` so future invocations distinguish this PRD from in-progress drafts. Record finalization to `.decision-log.md`. Share artifact paths. Common next: `bmad-ux`, `bmad-architecture`, `bmad-create-epics-and-stories`; invoke `bmad-help` for authoritative routing.
 8. Run `{workflow.on_complete}` if non-empty.

src/bmm-skills/2-plan-workflows/bmad-prd/assets/headless-schemas.md

@@ -18,7 +18,7 @@ Every headless run ends with one of these payloads. Omit keys for artifacts not
   "intent": "create",
   "prd": "{doc_workspace}/prd.md",
   "addendum": "{doc_workspace}/addendum.md",
-  "memlog": "{doc_workspace}/.memlog.md",
+  "decision_log": "{doc_workspace}/.decision-log.md",
   "open_questions": [],
   "assumptions": [],
   "external_handoffs": [
@@ -34,7 +34,7 @@ Every headless run ends with one of these payloads. Omit keys for artifacts not
   "status": "complete",
   "intent": "update",
   "prd": "{doc_workspace}/prd.md",
-  "memlog": "{doc_workspace}/.memlog.md",
+  "decision_log": "{doc_workspace}/.decision-log.md",
   "changes_summary": "1-3 sentences describing what changed and why",
   "conflicts_with_prior_decisions": [],
   "open_questions": [],

src/bmm-skills/2-plan-workflows/bmad-prd/customize.toml

@@ -60,7 +60,7 @@ validation_checklist_template = "assets/prd-validation-checklist.md"
 # collapse — no JS.
 validation_report_template = "assets/validation-report-template.html"
 
-# Run folder location. The PRD, optional addendum, memlog, and optional
+# Run folder location. The PRD, optional addendum, decision log, and optional
 # validation report all land inside `{prd_output_path}/{run_folder_pattern}/`.
 # Resume-check scans `{prd_output_path}` for prior unfinished runs.
 prd_output_path = "{planning_artifacts}/prds"

src/bmm-skills/2-plan-workflows/bmad-prd/references/headless.md

@@ -34,6 +34,6 @@ End with the JSON response (full schemas with examples in `assets/headless-schem
 
 ## Mode-specific overrides
 
-**Update.** Apply the change, log it via `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type change --text "<change + rationale>"`, and surface any conflict-with-prior-decision in `conflicts_with_prior_decisions[]` in the JSON status. Halt `blocked` if intent is ambiguous.
+**Update.** Apply the change, log to `.decision-log.md` with rationale, and surface any conflict-with-prior-decision in `conflicts_with_prior_decisions[]` in the JSON status. Halt `blocked` if intent is ambiguous.
 
 **Validate.** Always write both `validation-report.html` and `validation-report.md` to `{doc_workspace}` regardless of finding count. Always include `"offer_to_update": true` in the JSON status. Skip the browser-open step in `references/validate.md` — write the artifacts and return.

src/bmm-skills/2-plan-workflows/bmad-prd/references/validate.md

@@ -4,7 +4,7 @@ The Validate intent playbook. Standalone — this intent critiques an existing P
 
 ## Orient
 
-Source-extract against `.memlog.md`, any original inputs, and the PRD/addendum themselves. Delegate to subagents per PRD Discipline → "Extract, don't ingest" (in SKILL.md); the parent assembles from extracts.
+Source-extract against `.decision-log.md`, any original inputs, and the PRD/addendum themselves. Delegate to subagents per PRD Discipline → "Extract, don't ingest" (in SKILL.md); the parent assembles from extracts.
 
 ## Run the Reviewer Gate
 

src/bmm-skills/2-plan-workflows/bmad-ux/SKILL.md

@@ -30,7 +30,7 @@ UX may lead, follow, or stand alone. Inherit `sources:` by reference; the spines
 
 ## On Activation
 
-1. Resolve customization: `uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly and use defaults.
+1. Resolve customization: `python3 {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly and use defaults.
 2. Run `{workflow.activation_steps_prepend}`. Treat `{workflow.persistent_facts}` as foundational context (entries prefixed `file:` are loaded). `{workflow.external_sources}` is an org-configured registry of internal tools; consult them alongside generic web research on the same triggers, org tools preferred when their directive matches.
 3. Load `{project-root}/_bmad/bmm/config.yaml` (+ `config.user.yaml` if present). Resolve `{user_name}`, `{communication_language}`, `{document_output_language}`, `{planning_artifacts}`, `{project_name}`, `{date}`. Missing keys → neutral defaults; never block.
 4. If headless, follow `references/headless.md` for the whole run. Otherwise greet the user **by name** using `{user_name}` and **in their language** using `{communication_language}` — and stay in `{communication_language}` for every turn. In the greeting, let the user know `bmad-party-mode` and `bmad-advanced-elicitation` are always available. Then scan for misroute on the first message: PRD → `bmad-prd`; architecture → `bmad-architecture`; game UX → BMad GDS; agent/skill → `bmad-workflow-builder`; brief → `bmad-product-brief`.
@@ -42,15 +42,15 @@ Activation is complete. If `activation_steps_prepend` or `activation_steps_appen
 
 ## Modes
 
-**Create.** Bind `{doc_workspace}` to `{workflow.ux_output_path}/{workflow.run_folder_pattern}/`. Create `.working/` and `imports/`; seed the memlog with `uv run {project-root}/_bmad/scripts/memlog.py init --workspace {doc_workspace} --field topic="<product/UX>"`; create `DESIGN.md` (frontmatter only) and `EXPERIENCE.md` (frontmatter only). Run Discovery → Finalize.
+**Create.** Bind `{doc_workspace}` to `{workflow.ux_output_path}/{workflow.run_folder_pattern}/`. Create `.working/`, `imports/`, `.decision-log.md`, `DESIGN.md` (frontmatter only), and `EXPERIENCE.md` (frontmatter only). Run Discovery → Finalize.
 
-**Update.** Read spines + memlog + sources. If `.memlog.md` is missing, init it with `uv run {project-root}/_bmad/scripts/memlog.py init --workspace {doc_workspace}` — this update is entry one. Surface conflicts with prior decisions. Run Finalize.
+**Update.** Read spines + log + sources. Create the log if missing — this update is entry one. Surface conflicts with prior decisions. Run Finalize.
 
 **Validate.** See `references/validate.md`.
 
 ## Discovery
 
-**Capture; do not author.** The spines are distilled at Finalize toward the memlog. Decisions → `.memlog.md` (canonical), each appended via `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type <decision|change|override|assumption|event> --text "…"` — never hand-edited; a resume reloads it. Creative-tool artifacts → `.working/`. User-supplied visuals (Figma, sketches, brand decks, image folders) → `imports/`, one `memlog.py append` per item. Spines win on conflict.
+**Capture; do not author.** The spines are distilled at Finalize. Decisions → `.decision-log.md` (canonical). Creative-tool artifacts → `.working/`. User-supplied visuals (Figma, sketches, brand decks, image folders) → `imports/`, one log line per item. Spines win on conflict.
 
 **Source scan.** Glob `{planning_artifacts}/` for candidate input paths; surface paths only — never read content in the parent. User confirms which apply or adds others; subagent-extracts on confirm.
 
@@ -80,11 +80,11 @@ Used by Validate and Finalize. **Opt-in, lens-selectable** — reviewers are cos
 
 Outcomes, in order:
 
-- **Spines distilled.** Subagent reads `.memlog.md`, `.working/`, `imports/`, sources; produces `DESIGN.md` against `## The DESIGN.md spine` + `{workflow.design_md_examples}` and `EXPERIENCE.md` against `## The EXPERIENCE.md spine` + `{workflow.experience_md_examples}`. Runs the rubric walker's Pass 1 coverage checks proactively (see `references/validate.md`). Surface gaps; never invent.
+- **Spines distilled.** Subagent reads `.decision-log.md`, `.working/`, `imports/`, sources; produces `DESIGN.md` against `## The DESIGN.md spine` + `{workflow.design_md_examples}` and `EXPERIENCE.md` against `## The EXPERIENCE.md spine` + `{workflow.experience_md_examples}`. Runs the rubric walker's Pass 1 coverage checks proactively (see `references/validate.md`). Surface gaps; never invent.
 - **Inputs reconciled.** Subagent per user-supplied input → `reconcile-{slug}.md`. Surface dropped qualitative ideas.
 - **Reviewer Gate offered.** Ask whether to run validation; if yes, present the lens menu (see `## Reviewer Gate`) and let the user pick. If any lens ran, resolve findings before polish; otherwise proceed.
-- **Open items triaged.** Open Questions, `[ASSUMPTION]`, `[NOTE FOR UX]`. Phase-blockers one at a time; non-blockers → `memlog.py append`.
+- **Open items triaged.** Open Questions, `[ASSUMPTION]`, `[NOTE FOR UX]`. Phase-blockers one at a time; non-blockers → log.
 - **Key-screen mocks rendered.** Key-screens tool → `.working/` for surfaces where layout drives behavior or anchors visual language.
 - **Mock coverage confirmed.** Walk every IA surface; classify *mocked* vs *spine-only*. Ask: *"These will be built from spine tables alone — any need a visual reference?"* Render more if named; log spine-only choices.
 - **Layout extracted, artifacts promoted.** Distill subagent re-reads each `.working/` and `imports/` artifact; lifts visual decisions into DESIGN.md and behavioral decisions into EXPERIENCE.md. Promote `.working/` keepers to `mockups/` (HTML) or `wireframes/` (Excalidraw); imports stay. Inline relative links at relevant spine sections; state spines-win-on-conflict once.
-- **Polished, handed off, closed.** Apply `{workflow.doc_standards}` in order. Execute `{workflow.external_handoffs}`; surface URLs. Set both files' `status: final`, `updated: {date}`. Log finalization via `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type event --text "spines finalized"`. Share paths. Common next: `bmad-architecture`, `bmad-create-epics-and-stories`, `bmad-dev-story`. Run `{workflow.on_complete}`.
+- **Polished, handed off, closed.** Apply `{workflow.doc_standards}` in order. Execute `{workflow.external_handoffs}`; surface URLs. Set both files' `status: final`, `updated: {date}`. Log finalization. Share paths. Common next: `bmad-architecture`, `bmad-create-epics-and-stories`, `bmad-dev-story`. Run `{workflow.on_complete}`.

src/bmm-skills/2-plan-workflows/bmad-ux/assets/design-directions.md

@@ -4,6 +4,6 @@ Subagent prompt. Produce 3-6 distinct visual directions for the product's hero s
 
 Each direction is a *complete visual personality* applied to the same key screen — not a palette swap. Differ on density, type weight, motion implication, brand register. Each file: 2-3 sentence rationale, near-1:1 hero screen mockup in a phone or browser frame, ideally a secondary screen, at least one state variant visible (aging row, empty state, etc).
 
-Use real product content from the conversation. Voice/tone from `.memlog.md` applied to every visible string — no lorem. Inline CSS, system fonts, no JS or network. Document hex values in `<style>` comments per direction.
+Use real product content from the conversation. Voice/tone from `.decision-log.md` applied to every visible string — no lorem. Inline CSS, system fonts, no JS or network. Document hex values in `<style>` comments per direction.
 
 Return to the parent: file paths, one-line personality summary per direction, what hero screen was depicted. Do not dump HTML into parent context. If interactive, open each file in the browser.

src/bmm-skills/2-plan-workflows/bmad-ux/assets/headless-schemas.md

@@ -18,7 +18,7 @@ Every headless run ends with one of these payloads. Omit keys for artifacts not
   "intent": "create",
   "design": "{doc_workspace}/DESIGN.md",
   "experience": "{doc_workspace}/EXPERIENCE.md",
-  "memlog": "{doc_workspace}/.memlog.md",
+  "decision_log": "{doc_workspace}/.decision-log.md",
   "working_artifacts": ["{doc_workspace}/.working/color-themes-1.html"],
   "promoted_artifacts": {
     "mockups": ["{doc_workspace}/mockups/direction-calm-sage.html"],
@@ -42,7 +42,7 @@ The `working_artifacts` and `promoted_artifacts` keys are optional and omitted e
   "intent": "update",
   "design": "{doc_workspace}/DESIGN.md",
   "experience": "{doc_workspace}/EXPERIENCE.md",
-  "memlog": "{doc_workspace}/.memlog.md",
+  "decision_log": "{doc_workspace}/.decision-log.md",
   "changes_summary": "1-3 sentences describing what changed and why",
   "conflicts_with_prior_decisions": [],
   "open_questions": [],

src/bmm-skills/2-plan-workflows/bmad-ux/assets/key-screens.md

@@ -4,11 +4,11 @@ Subagent prompt. Fired at Finalize (or during late Discovery once layout decisio
 
 ## Inputs
 
-`.memlog.md`, the current drafts `DESIGN.md` and `EXPERIENCE.md`, `.working/` (especially the chosen color-theme and direction mocks), source PRD. The user names which surfaces to render — typically 2-4: the canonical entry surface, the most complex flow's hero screen, any load-bearing overlay/modal, and (when present) the Week / list / dashboard view.
+`.decision-log.md`, the current drafts `DESIGN.md` and `EXPERIENCE.md`, `.working/` (especially the chosen color-theme and direction mocks), source PRD. The user names which surfaces to render — typically 2-4: the canonical entry surface, the most complex flow's hero screen, any load-bearing overlay/modal, and (when present) the Week / list / dashboard view.
 
 ## What to render
 
-One HTML file per screen, at `.working/key-{slug}.html`. Each file: realistic device frame (phone or browser), real product content from the conversation (no lorem), every visible string voice-checked against `.memlog.md`, all decided tokens applied. Show one canonical state per screen; if a surface has a load-bearing alternate state (focus, error, crisis-card-present), render it as a second column or section in the same file.
+One HTML file per screen, at `.working/key-{slug}.html`. Each file: realistic device frame (phone or browser), real product content from the conversation (no lorem), every visible string voice-checked against `.decision-log.md`, all decided tokens applied. Show one canonical state per screen; if a surface has a load-bearing alternate state (focus, error, crisis-card-present), render it as a second column or section in the same file.
 
 Inline CSS, system fonts, no JS, no network. The mock must render fully offline. Comment block at the top of the `<style>` notes which spine sections govern this screen so a future reader knows what to check.
 
@@ -23,7 +23,7 @@ The parent, at Finalize "Promote working artifacts," uses this summary to insert
 
 ## Anti-patterns
 
-- Do not invent layout — every composition decision must trace to a `.working/` artifact or a confirmation in `.memlog.md`. If a layout question is open, the mock is premature.
+- Do not invent layout — every composition decision must trace to a `.working/` artifact or a confirmation in `.decision-log.md`. If a layout question is open, the mock is premature.
 - Do not show every screen of every flow — 2-4 load-bearing surfaces, not 14.
-- Do not stage marketing copy. Strings come from `.memlog.md` and voice rules.
+- Do not stage marketing copy. Strings come from `.decision-log.md` and voice rules.
 - Do not introduce a new pattern not in the spine's Component Patterns table. If you need one, log it and ask before rendering.

src/bmm-skills/2-plan-workflows/bmad-ux/customize.toml

@@ -53,7 +53,7 @@ design_handoffs = [
 # HTML skeleton filled in by the validation synthesis pass.
 validation_report_template = "assets/validation-report-template.html"
 
-# Run folder. DESIGN.md, EXPERIENCE.md, .memlog.md, .working/
+# Run folder. DESIGN.md, EXPERIENCE.md, .decision-log.md, .working/
 # (creative-tool artifacts), imports/ (user-supplied screens / brand decks /
 # Figma exports / sketches), optional mockups/ and wireframes/ (promoted
 # artifacts), optional validation-report.* all land inside

src/bmm-skills/2-plan-workflows/bmad-ux/references/creative-tools.md

@@ -14,6 +14,6 @@ Every renderer writes to `{doc_workspace}/.working/` with a descriptive filename
 
 ## Renderer contract
 
-The parent passes the subagent: current `.memlog.md`, relevant prior `.working/` captures, the user's stated intent for this pass, the output path. The subagent writes its artifact under `.working/` and returns ONLY a compact summary (file path, one line per variant, mode coverage). Parent never holds the full payload.
+The parent passes the subagent: current `.decision-log.md`, relevant prior `.working/` captures, the user's stated intent for this pass, the output path. The subagent writes its artifact under `.working/` and returns ONLY a compact summary (file path, one line per variant, mode coverage). Parent never holds the full payload.
 
 For HTML, open in browser when interactive: `python3 -c "import webbrowser, pathlib; webbrowser.open(pathlib.Path('PATH').resolve().as_uri())"`. Skip in headless.

src/bmm-skills/2-plan-workflows/bmad-ux/references/headless.md

@@ -32,6 +32,6 @@ End with JSON matching `assets/headless-schemas.md`. `intent` reflects detected
 
 ## Mode-specific overrides
 
-**Update.** Apply the change. Log it via `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type change --text "<change + rationale>"`. Surface conflicts in `conflicts_with_prior_decisions[]`.
+**Update.** Apply the change. Log to `.decision-log.md` with rationale. Surface conflicts in `conflicts_with_prior_decisions[]`.
 
 **Validate.** Always write both `validation-report.html` and `validation-report.md` regardless of finding count. Always include `"offer_to_update": true`. Skip the browser-open step.

src/bmm-skills/2-plan-workflows/bmad-ux/references/validate.md

@@ -4,7 +4,7 @@ Critique an existing spine pair (`DESIGN.md` + `EXPERIENCE.md`) or any format of
 
 ## Orient
 
-Subagent-extract from `.memlog.md`, sources in frontmatter, `imports/`, `mockups/`, `wireframes/`, `DESIGN.md`, `EXPERIENCE.md`. Parent assembles from extracts.
+Subagent-extract from `.decision-log.md`, sources in frontmatter, `imports/`, `mockups/`, `wireframes/`, `DESIGN.md`, `EXPERIENCE.md`. Parent assembles from extracts.
 
 ## Reviewer Gate
 
@@ -34,7 +34,7 @@ Rubric walker prompt:
 >
 > 7. **Inheritance discipline.** `sources` frontmatter resolves. UJ / requirement names verbatim from sources. Glossary identical across spines and sources. Component names identical across all sections in both files. EXPERIENCE.md token references resolve to DESIGN.md tokens by name.
 >
-> 8. **Shape fit.** DESIGN.md sections in canonical order (Brand & Style → Colors → Typography → Layout & Spacing → Elevation & Depth → Shapes → Components → Do's and Don'ts; omittable but order-locked when present). EXPERIENCE.md required defaults present (Foundation, IA, Voice and Tone, Component Patterns, State Patterns, Interaction Primitives, Accessibility Floor, Key Flows). Dropped defaults defensible. Required-when-applicable present where triggered (Inspiration when sources / memlog show reference products or rejects; Responsive when multi-surface or breakpoints). Invented sections earn their place.
+> 8. **Shape fit.** DESIGN.md sections in canonical order (Brand & Style → Colors → Typography → Layout & Spacing → Elevation & Depth → Shapes → Components → Do's and Don'ts; omittable but order-locked when present). EXPERIENCE.md required defaults present (Foundation, IA, Voice and Tone, Component Patterns, State Patterns, Interaction Primitives, Accessibility Floor, Key Flows). Dropped defaults defensible. Required-when-applicable present where triggered (Inspiration when sources / log show reference products or rejects; Responsive when multi-surface or breakpoints). Invented sections earn their place.
 >
 > Severity = downstream impact, not fix difficulty.
 >

src/bmm-skills/3-solutioning/bmad-architecture/SKILL.md

@@ -32,12 +32,12 @@ The input itself tells you what kind of job this is — read it rather than quiz
 
 ## How a run works
 
-The **memlog** (`.memlog.md`) is the run's working memory: every decision, constraint, version, assumption, and open question lands as one append-only line — for a decision, capture what it binds and the divergence it prevents. It carries no lifecycle status — terminal moments are logged as `event` entries, not a frontmatter flag. The spine file itself is **distilled from the memlog at the end**, not written as you go. Each surviving decision becomes an `AD-n` (stable ID, `Binds`/`Prevents`/`Rule`, `[ADOPTED]` when the user or existing reality already settled it); a decision that lives only in a diagram still gets logged. Resume a prior run by reloading its memlog.
+The **memlog** (`.memlog.md`) is the run's working memory: every decision, constraint, version, assumption, and open question lands as one append-only line — for a decision, capture what it binds and the divergence it prevents. It is the shared canonical memlog (the same `{project-root}/_bmad/scripts/memlog.py` bmad-spec writes through), so it carries no lifecycle status — terminal moments are logged as `event` entries, not a frontmatter flag. The spine is **distilled from the memlog at the end**, not written as you go. Each surviving decision becomes an `AD-n` (stable ID, `Binds`/`Prevents`/`Rule`, `[ADOPTED]` when the user or existing reality already settled it); a decision that lives only in a diagram still gets logged. Resume a prior run by reloading its memlog.
 
 Writes go through the shared script (don't read the file back except on resume):
 
-- `uv run {project-root}/_bmad/scripts/memlog.py init --workspace {doc_workspace} --field scope="…" --field purpose="…" --field altitude="…"`
-- `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type <decision|constraint|version|assumption|question|direction|event> --text "…"`
+- `python3 {project-root}/_bmad/scripts/memlog.py init --workspace {doc_workspace} --field scope="…" --field purpose="…" --field altitude="…"`
+- `python3 {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type <decision|constraint|version|assumption|question|direction|event> --text "…"`
 
 ## Resolution rules
 
@@ -50,7 +50,7 @@ Writes go through the shared script (don't read the file back except on resume):
 
 **Forwarded activation:** if a caller (e.g. the `bmad-create-architecture` shim) invoked you with a stated intent and pre-resolved customization fields, honor them verbatim — skip your own intent inference, use the supplied values for those named fields, and resolve only the remaining fields from your own `customize.toml`.
 
-1. Resolve customization: `uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow` (on failure read `{skill-root}/customize.toml`, use defaults). Run `{workflow.activation_steps_prepend}`, then `{workflow.activation_steps_append}`. Hold `{workflow.persistent_facts}` as standing context — the default loads `project-context.md`, load-bearing for brownfield — and consult `{workflow.external_sources}` on demand.
+1. Resolve customization: `python3 {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow` (on failure read `{skill-root}/customize.toml`, use defaults). Run `{workflow.activation_steps_prepend}`, then `{workflow.activation_steps_append}`. Hold `{workflow.persistent_facts}` as standing context — the default loads `project-context.md`, load-bearing for brownfield — and consult `{workflow.external_sources}` on demand.
 2. Load `{project-root}/_bmad/bmm/config.yaml` (+ `config.user.yaml`) for `{user_name}`, `{communication_language}`, `{document_output_language}`, `{planning_artifacts}`, `{project_name}`, `{date}`; missing keys take neutral defaults, never block.
 3. Headless (no interactive user) → follow `references/headless.md` for the whole run. Otherwise greet `{user_name}` in `{communication_language}`. Detect the intent from the conversation and input — **create** (the default), **update** an existing spine, or **validate** one (see those sections). If the real ask is requirements / UX / a capability contract / epic breakdown / an agent, invoke the `bmad-prd`, `bmad-ux`, `bmad-spec`, `bmad-create-epics-and-stories`, or `bmad-workflow-builder` (if the BMad Builder module is installed) skill instead.
 4. If a run folder for this target already exists under `{workflow.spine_output_path}`, offer to resume from its memlog rather than restart.

src/bmm-skills/3-solutioning/bmad-architecture/references/reviewer-gate.md

@@ -2,7 +2,7 @@
 
 The spine's pre-handoff review. Runs at Finalize (after distill + reconcile) and *is* the Validate intent. The difference is the ending: at Finalize you apply the clear fixes yourself; under Validate you report and don't change the spine.
 
-Cheap deterministic pass first: `uv run {skill-root}/scripts/lint_spine.py --workspace {doc_workspace}` settles the mechanical misses (placeholders, duplicate `AD` IDs, missing Binds/Prevents/Rule, unpinned Stack versions), so reviewers spend judgment on the semantic half.
+Cheap deterministic pass first: `python3 {skill-root}/scripts/lint_spine.py --workspace {doc_workspace}` settles the mechanical misses (placeholders, duplicate `AD` IDs, missing Binds/Prevents/Rule, unpinned Stack versions), so reviewers spend judgment on the semantic half.
 
 Assemble the menu: a **rubric walker** that judges the spine against the good-spine checklist below, **+ every entry in `{workflow.finalize_reviewers}`**, + ad-hoc lenses you invent or offer as the spine's rigor, altitude, and criticality warrant — a security/compliance lens for regulated stakes, a seam reviewer cross-team, a data-integrity lens for a heavy data model. Scale *whether and how heavily the gate runs* to the stakes: a throwaway prototype may run it quietly or skip the gate entirely; a high-criticality or platform-altitude spine earns more lenses and the explicit all / subset / skip menu. But once the gate runs, the `{workflow.finalize_reviewers}` always run — they are the configured floor, never cherry-picked out; only the ad-hoc lenses are optional. (Headless never skips the gate.)
 

src/bmm-skills/4-implementation/bmad-dev-auto/SKILL.md

@@ -1,101 +0,0 @@
----
-name: bmad-dev-auto
-description: 'One iteration of an unattended development loop. Use when invoked by name.'
----
-
-# Dev Auto Workflow
-
-**Goal:** Turn intent into a hardened, reviewable artifact, without human interaction.
-
-**CRITICAL:** If a step says "read fully and follow step-XX", you read and follow step-XX. No exceptions.
-
-## HALT
-
-To HALT with a final status and optional blocking condition:
-
-1. If `{spec_file}` is known and exists, update `status` in frontmatter and append missing result details under `## Auto Run Result`.
-2. If `{spec_file}` is unknown or missing, create `{implementation_artifacts}/bmad-dev-auto-result-<slug-or-timestamp>.md` with:
-   ```markdown
-   ---
-   status: <final status>
-   ---
-
-   # BMad Dev Auto Result
-
-   Status: <final status>
-   Blocking condition: <blocking condition, if any>
-   ```
-3. Run: `python3 {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow.on_complete`
-4. If the resolved `workflow.on_complete` is non-empty, follow it as the final instruction before exiting.
-5. Stop the workflow.
-
-## Subagents
-
-Using subagents when instructed is mandatory. If you cannot, HALT with status `blocked` and blocking condition `no subagents`.
-
-## READY FOR DEVELOPMENT STANDARD
-
-A specification is "Ready for Development" when:
-
-- **Actionable**: Every task has a file path and specific action.
-- **Logical**: Tasks ordered by dependency.
-- **Testable**: All ACs use Given/When/Then.
-- **Complete**: No placeholders or TBDs.
-- **Sufficient**: No known requirement, acceptance, dependency, or implementation gaps remain unresolved.
-- **Coherent**: No unresolved ambiguities or internal contradictions.
-
-## Conventions
-
-- Bare paths (e.g. `step-01-clarify-and-route.md`) resolve from the skill root.
-- `{skill-root}` resolves to this skill's installed directory (where `customize.toml` lives).
-- `{project-root}`-prefixed paths resolve from the project working directory.
-- `{skill-name}` resolves to the skill directory's basename.
-
-## On Activation
-
-### Step 1: Resolve the Workflow Block
-
-Run: `python3 {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`
-
-**If the script fails**, resolve the `workflow` block yourself by reading these three files in base → team → user order and applying the same structural merge rules as the resolver:
-
-1. `{skill-root}/customize.toml` — defaults
-2. `{project-root}/_bmad/custom/{skill-name}.toml` — team overrides
-3. `{project-root}/_bmad/custom/{skill-name}.user.toml` — personal overrides
-
-Any missing file is skipped. Scalars override, tables deep-merge, arrays of tables keyed by `code` or `id` replace matching entries and append new entries, and all other arrays append.
-
-### Step 2: Execute Prepend Steps
-
-Execute each entry in `{workflow.activation_steps_prepend}` in order before proceeding.
-
-### Step 3: Load Persistent Facts
-
-Treat every entry in `{workflow.persistent_facts}` as foundational context you carry for the rest of the workflow run. Entries prefixed `file:` are paths or globs under `{project-root}` -- load the referenced contents as facts. All other entries are facts verbatim.
-
-### Step 4: Load Config
-
-Load config from `{project-root}/_bmad/bmm/config.yaml` and resolve:
-
-- `project_name`, `planning_artifacts`, `implementation_artifacts`, `user_name`
-- `communication_language`, `document_output_language`, `user_skill_level`
-- `date` as system-generated current datetime
-- `project_context` = `**/project-context.md` (load if exists)
-- CLAUDE.md / memory files (load if exist)
-- YOU MUST ALWAYS SPEAK OUTPUT in your Agent communication style with the config `{communication_language}`
-- Language MUST be tailored to `{user_skill_level}`
-- Generate all documents in `{document_output_language}`
-
-### Step 5: Execute Append Steps
-
-Execute each entry in `{workflow.activation_steps_append}` in order.
-
-Activation is complete after all activation steps have run.
-
-## Workflow Execution
-
-Follow the step files in order. Read one step fully, execute it, then load the next step only when directed. Do not skip, reorder, or pre-load steps.
-
-## First workflow step
-
-Read fully and follow: `./step-01-clarify-and-route.md` to begin the workflow.

src/bmm-skills/4-implementation/bmad-dev-auto/compile-epic-context.md

@@ -1,62 +0,0 @@
-# Compile Epic Context
-
-**Task**
-Given an epic number, the epics file, the planning artifacts directory, and a desired output path, compile a clean, focused, developer-ready context file (`epic-<N>-context.md`).
-
-**Steps**
-
-1. Read the epics file and extract the target epic's title, goal, and list of stories.
-2. Scan the planning artifacts directory for the standard files (PRD, architecture, UX/design, product brief).
-3. Pull only the information relevant to this epic.
-4. Write the compiled context to the exact output path using the format below.
-
-## Exact Output Format
-
-Use these headings:
-
-```markdown
-# Epic {N} Context: {Epic Title}
-
-<!-- Generated from planning artifacts. Regenerate with compile-epic-context if planning docs change. -->
-
-## Goal
-
-{One clear paragraph: what this epic achieves and why it matters.}
-
-## Stories
-
-- Story X.Y: Brief title only
-- ...
-
-## Requirements & Constraints
-
-{Relevant functional/non-functional requirements and success criteria for this epic (describe by purpose, not source).}
-
-## Technical Decisions
-
-{Key architecture decisions, constraints, patterns, data models, and conventions relevant to this epic.}
-
-## UX & Interaction Patterns
-
-{Relevant UX flows, interaction patterns, and design constraints (omit section entirely if nothing relevant).}
-
-## Cross-Story Dependencies
-
-{Dependencies between stories in this epic or with other epics/systems (omit if none).}
-```
-
-## Rules
-
-- **Scope aggressively.** Include only what a developer working on any story in this epic actually needs. When in doubt, leave it out — the developer can always read the full planning doc.
-- **Describe by purpose, not by source.** Write "API responses must include pagination metadata" not "Per PRD section 3.2.1, pagination is required." Planning doc internals will change; the constraint won't.
-- **No full copies.** Never quote source documents, section numbers, or paste large blocks verbatim. Always distill.
-- **No story-level details.** The story list is for orientation only. Individual story specs handle the details.
-- **Nothing derivable from the codebase.** Don't document what a developer can learn by reading the code.
-- **Be concise and actionable.** Target 800–1500 tokens total. This file loads into bmad-dev-auto's context alongside other material.
-- **Never hallucinate content.** If source material doesn't say something, don't invent it.
-- **Omit empty sections entirely**, except Goal and Stories, which are always required.
-
-## Error handling
-
-- **If the epics file is missing or the target epic is not found:** write nothing and report the problem to the calling agent. Goal and Stories cannot be populated without a usable epics file.
-- **If planning artifacts are missing or empty:** still produce the file with Goal and Stories populated from the epics file. Under Requirements & Constraints, write: "Planning artifacts were unavailable; only epics-file context was used." Never hallucinate content to fill missing sections.

src/bmm-skills/4-implementation/bmad-dev-auto/customize.toml

@@ -1,33 +0,0 @@
-# DO NOT EDIT -- overwritten on every update.
-#
-# Default customization values for bmad-dev-auto.
-# Override in _bmad/custom/bmad-dev-auto.toml or
-# _bmad/custom/bmad-dev-auto.user.toml.
-#
-# Merge rules:
-# - Strings replace the default.
-# - Lists append to the default list.
-# - Tables merge key by key.
-
-[workflow]
-
-# Extra instructions to run before config is loaded.
-
-activation_steps_prepend = []
-
-# Extra instructions to run after config is loaded and before step 01.
-
-activation_steps_append = []
-
-# Facts kept in context for the whole run.
-# Entries are literal text or file references prefixed with "file:".
-# File entries may use globs and are loaded during activation.
-
-persistent_facts = [
-  "file:{project-root}/**/project-context.md",
-]
-
-# Instruction run by HALT after writing the terminal result.
-# Empty means no extra terminal behavior.
-
-on_complete = ""

src/bmm-skills/4-implementation/bmad-dev-auto/spec-template.md

@@ -1,90 +0,0 @@
----
-title: '{title}'
-type: 'feature' # feature | bugfix | refactor | chore
-created: '{date}'
-status: 'draft' # draft | ready-for-dev | in-progress | in-review | done | blocked
-review_loop_iteration: 0 # incremented by step-04 before each review loopback
-context: [] # optional: `{project-root}/`-prefixed paths to project-wide standards/docs the implementation agent should load. Keep short — only what isn't already distilled into the spec body.
-warnings: [] # optional: machine-readable warnings for orchestration, e.g. oversized, multiple-goals
----
-
-<!-- Aim for 900–1600 tokens. If larger, add `oversized` to frontmatter `warnings` and continue.
-     Never over-specify "how" — use boundaries + examples instead.
-     Cohesive cross-layer stories (DB+BE+UI) stay in ONE file.
-     IMPORTANT: Remove all HTML comments when filling this template. -->
-
-<intent-contract>
-
-## Intent
-
-<!-- What is broken or missing, and why it matters. Then the high-level approach — the "what", not the "how". -->
-
-**Problem:** ONE_TO_TWO_SENTENCES
-
-**Approach:** ONE_TO_TWO_SENTENCES
-
-## Boundaries & Constraints
-
-<!-- Three tiers: Always = invariant rules. Block If = decisions that cannot be made unattended. Never = out of scope + forbidden approaches. -->
-
-**Always:** INVARIANT_RULES
-
-**Block If:** DECISIONS_REQUIRING_HUMAN_INPUT
-<!-- Agent: if any of these trigger during execution, HALT with status blocked and the blocking condition. -->
-
-**Never:** NON_GOALS_AND_FORBIDDEN_APPROACHES
-
-## I/O & Edge-Case Matrix
-
-<!-- If no meaningful I/O scenarios exist, DELETE THIS ENTIRE SECTION. Do not write "N/A" or "None". -->
-
-| Scenario | Input / State | Expected Output / Behavior | Error Handling |
-|----------|--------------|---------------------------|----------------|
-| HAPPY_PATH | INPUT | OUTCOME | No error expected |
-| ERROR_CASE | INPUT | OUTCOME | ERROR_HANDLING |
-
-</intent-contract>
-
-## Code Map
-
-<!-- Agent-populated during planning. Annotated paths prevent blind codebase searching. -->
-
-- `FILE` -- ROLE_OR_RELEVANCE
-- `FILE` -- ROLE_OR_RELEVANCE
-
-## Tasks & Acceptance
-
-<!-- Tasks: backtick-quoted file path -- action -- rationale. Prefer one task per file; group tightly-coupled changes when splitting would be artificial. -->
-<!-- If an I/O Matrix is present, include a task to unit-test its edge cases. -->
-<!-- AC covers system-level behaviors not captured by the I/O Matrix. Do not duplicate I/O scenarios here. -->
-
-**Execution:**
-- [ ] `FILE` -- ACTION -- RATIONALE
-
-**Acceptance Criteria:**
-- Given PRECONDITION, when ACTION, then EXPECTED_RESULT
-
-## Spec Change Log
-
-<!-- Append-only. Populated by step-04 during review loops. Do not modify or delete existing entries.
-     Each entry records: what finding triggered the change, what was amended, what known-bad state
-     the amendment avoids, and any KEEP instructions (what worked well and must survive re-derivation).
-     Empty until the first bad_spec loopback. -->
-
-## Design Notes
-
-<!-- If the approach is straightforward, DELETE THIS ENTIRE SECTION. Do not write "N/A" or "None". -->
-<!-- Design rationale and golden examples only when non-obvious. Keep examples to 5–10 lines. -->
-
-DESIGN_RATIONALE_AND_EXAMPLES
-
-## Verification
-
-<!-- If no build, test, or lint commands apply, DELETE THIS ENTIRE SECTION. Do not write "N/A" or "None". -->
-<!-- How the agent confirms its own work. Prefer CLI commands. When no CLI check applies, state what to inspect manually. -->
-
-**Commands:**
-- `COMMAND` -- expected: SUCCESS_CRITERIA
-
-**Manual checks (if no CLI):**
-- WHAT_TO_INSPECT_AND_EXPECTED_STATE

src/bmm-skills/4-implementation/bmad-dev-auto/step-01-clarify-and-route.md

@@ -1,66 +0,0 @@
----
-deferred_work_file: '{implementation_artifacts}/deferred-work.md'
-spec_file: '' # set at runtime for both routes before leaving this step
----
-
-# Step 1: Clarify and Route
-
-## RULES
-
-- YOU MUST ALWAYS SPEAK OUTPUT in your Agent communication style with the config `{communication_language}`
-- Treat the invocation intent as workflow input, not as a substitute for step-02 investigation and spec generation.
-- **EARLY EXIT** means: stop this step immediately, then read and follow the target file. Return here only if a later step explicitly says to loop back.
-
-## Intent check (do this first)
-
-Use the invocation prompt as the intent.
-
-If the invocation prompt explicitly points to an existing spec file with recognized `status` frontmatter, set `spec_file`, then **EARLY EXIT** to the appropriate step:
-- `draft` → `./step-02-plan.md`
-- `ready-for-dev` or `in-progress` → `./step-03-implement.md`
-- `in-review` → `./step-04-review.md`
-- `blocked` → HALT with status `blocked` and blocking condition `blocked spec supplied`.
-- `done` → ingest as context and proceed to INSTRUCTIONS — do not resume.
-
-Otherwise, treat the invocation prompt as starting intent. This may be a story ID, ticket ID, file path, short description, or longer free-form intent. Do not infer workflow state from non-spec files.
-If the invocation prompt does not contain enough intent to identify what to implement, HALT with status `blocked` and blocking condition `unclear intent`.
-
-## INSTRUCTIONS
-
-1. Load context.
-   - List files in `{planning_artifacts}` and `{implementation_artifacts}`.
-   - If the invocation prompt points to an unformatted spec or intent file, ingest that file. Do not scan for unrelated intent files.
-   - **Determine context strategy.** Using the intent and the artifact listing, infer whether the current work is a story from an epic. Do not rely on filename patterns or regex — reason about the intent, the listing, and any epics file content together.
-
-     **A) Epic story path** — if the intent is clearly an epic story:
-
-     1. Identify the epic number `{epic_num}` and (if present) the story number `{story_num}`. If you can't identify an epic number, use path B.
-
-     2. **Check for a valid cached epic context.** Look for `{implementation_artifacts}/epic-<N>-context.md` (where `<N>` is the epic number). A file is **valid** when it exists, is non-empty, starts with `# Epic <N> Context:` (with the correct epic number), and no file in `{planning_artifacts}` is newer.
-        - **If valid:** load it as the primary planning context. Do not load raw planning docs (PRD, architecture, UX, etc.).
-        - **If missing, empty, or invalid:** compile it in the next bullet.
-
-     3. **Compile epic context if needed.** If no valid cached epic context was loaded, produce `{implementation_artifacts}/epic-<N>-context.md` by spawning a subagent with `./compile-epic-context.md` as its prompt. Pass it the epic number, the epics file path, the `{planning_artifacts}` directory, and the output path `{implementation_artifacts}/epic-<N>-context.md`.
-
-     4. **Verify if compiled.** If epic context was compiled, verify the output file exists, is non-empty, and starts with `# Epic <N> Context:`. If valid, load it. If verification fails, HALT with status `blocked` and blocking condition `context compilation verification failed`.
-
-     5. **Previous story continuity.** Regardless of which context source succeeded above, scan `{implementation_artifacts}` for specs from the same epic with `status: done` and a lower story number. Load the most recent one (highest story number below current). Extract its **Code Map**, **Design Notes**, **Spec Change Log**, and **task list** as continuity context for step-02 planning. If no `done` spec is found but an `in-review` spec exists for the same epic with a lower story number, HALT with status `blocked` and blocking condition `missing previous-story continuity decision`.
-
-     **B) Freeform path** — if the intent is not an epic story:
-     - Planning artifacts are the output of BMAD phases 1-3. Typical files include:
-       - **PRD** (`*prd*`) — product requirements and success criteria
-       - **Architecture** (`*architecture*`) — technical design decisions and constraints
-       - **UX/Design** (`*ux*`) — user experience and interaction design
-       - **Epics** (`*epic*`) — feature breakdown into implementable stories
-       - **Product Brief** (`*brief*`) — project vision and scope
-     - Scan the listing for files matching these patterns. If any look relevant to the current intent, load them selectively — you don't need all of them, but you need the right constraints and requirements rather than guessing from code alone.
-2. Resolve intent from the invocation prompt and loaded artifacts. Do not fantasize or leave open questions. If the intent cannot be resolved, HALT with status `blocked` and the unresolved questions as blocking condition.
-3. Version control sanity check. Is the working tree clean? Does the current branch make sense for this intent — considering its name and recent history? If the tree is dirty or the branch is an obvious mismatch, HALT with status `blocked` and that condition as blocking condition. If version control is unavailable, skip this check.
-4. Multi-goal warning. If the intent appears to contain multiple independently shippable goals, carry `multiple-goals` forward so step-02 can add it to `{spec_file}` frontmatter `warnings`. Do not split or block.
-5. Route:
-
-   Derive a valid kebab-case slug from the clarified intent. If the intent references a tracking identifier (story number, issue number, ticket ID), lead the slug with it (e.g. `3-2-digest-delivery`, `gh-47-fix-auth`). If `{implementation_artifacts}/spec-{slug}.md` already exists: if its status is `draft`, treat it as the same work and resume it (set `spec_file` to that path, **EARLY EXIT** → `./step-02-plan.md`); otherwise append `-2`, `-3`, etc. Set `spec_file` = `{implementation_artifacts}/spec-{slug}.md`.
-
-## NEXT
-
-Read fully and follow `./step-02-plan.md`

src/bmm-skills/4-implementation/bmad-dev-auto/step-02-plan.md

@@ -1,32 +0,0 @@
----
-deferred_work_file: '{implementation_artifacts}/deferred-work.md'
----
-
-# Step 2: Plan
-
-## RULES
-
-- YOU MUST ALWAYS SPEAK OUTPUT in your Agent communication style with the config `{communication_language}`
-- No human interaction: do not ask questions or wait for approval in this step.
-
-## INSTRUCTIONS
-
-1. Draft resume check. If `{spec_file}` exists with `status: draft`, read it and capture the verbatim `<intent-contract>...</intent-contract>` block as `preserved_intent_contract`. Otherwise `preserved_intent_contract` is empty.
-2. Investigate codebase. _Use subagents for deep exploration. To prevent context snowballing, instruct subagents to give you distilled summaries only._
-3. Read `./spec-template.md` fully. Fill it out based on the intent and investigation. If `{preserved_intent_contract}` is non-empty, substitute it for the `<intent-contract>` block in your filled spec before writing. Write the result to `{spec_file}`.
-4. Self-review against READY FOR DEVELOPMENT standard.
-5. If intent gaps exist, do not fantasize and do not leave open questions. HALT with status `blocked`, blocking condition `intent gaps`, and include the unanswered questions and evidence gathered.
-6. Warning check. If step-01 carried `multiple-goals`, add it to `{spec_file}` frontmatter `warnings`. If `{spec_file}` exceeds 1600 tokens, add `oversized` to frontmatter `warnings`. Continue either way.
-
-### READY-FOR-DEVELOPMENT GATE
-
-Re-read `./SKILL.md`, then re-read `{spec_file}` from disk and verify the spec meets the READY FOR DEVELOPMENT standard.
-
-- **If the file is missing:** HALT with status `blocked` and blocking condition `planned spec file disappeared before implementation`.
-- **If the spec meets the standard:** set `{spec_file}` frontmatter status to `ready-for-dev`, then continue to step 3.
-- **If the spec does not meet the standard:** repair it once, then re-read it from disk and verify again. If it still does not meet the standard, HALT with status `blocked`, blocking condition `spec failed ready-for-development standard`, and include the failing criteria and evidence gathered.
-
-
-## NEXT
-
-Read fully and follow `./step-03-implement.md`

src/bmm-skills/4-implementation/bmad-dev-auto/step-03-implement.md

@@ -1,38 +0,0 @@
----
----
-
-# Step 3: Implement
-
-## RULES
-
-- YOU MUST ALWAYS SPEAK OUTPUT in your Agent communication style with the config `{communication_language}`
-- No human interaction: do not ask questions or wait for approval in this step.
-- Content inside `<intent-contract>` in `{spec_file}` is read-only. Do not modify.
-
-## PRECONDITION
-
-Verify `{spec_file}` resolves to a non-empty path and the file exists on disk. If empty or missing, HALT with status `blocked` and blocking condition `missing spec_file before implementation`.
-
-## INSTRUCTIONS
-
-### Baseline
-
-Capture `baseline_revision` (current HEAD, or `NO_VCS` if version control is unavailable) into `{spec_file}` frontmatter before making any changes.
-
-### Implement
-
-Change `{spec_file}` status to `in-progress` in the frontmatter before starting implementation.
-
-If `{spec_file}` has a non-empty `context:` list in its frontmatter, load those files before implementation begins. When handing to a subagent, include them in the subagent prompt so it has access to the referenced context.
-
-Hand `{spec_file}` to an implementation subagent.
-
-**Path formatting rule:** Any markdown links written into `{spec_file}` must use paths relative to `{spec_file}`'s directory so they are clickable in VS Code. Any file paths displayed in terminal/conversation output must use CWD-relative format with `:line` notation (e.g., `src/path/file.ts:42`) for terminal clickability. No leading `/` in either case.
-
-### Tasks & Acceptance Verification
-
-After the implementation subagent returns, verify every task in the `## Tasks & Acceptance` section of `{spec_file}` is complete and every acceptance criterion is satisfied. Mark each finished task `[x]`. If any task is not done or any acceptance criterion is not satisfied, finish the missing work before proceeding. If the missing work cannot be completed, HALT with status `blocked`, blocking condition `implementation verification failed`, and include the unfinished task or failing acceptance criterion and reason.
-
-## NEXT
-
-Read fully and follow `./step-04-review.md`

src/bmm-skills/4-implementation/bmad-dev-auto/step-04-review.md

@@ -1,61 +0,0 @@
----
-deferred_work_file: '{implementation_artifacts}/deferred-work.md'
----
-
-# Step 4: Review
-
-## RULES
-
-- YOU MUST ALWAYS SPEAK OUTPUT in your Agent communication style with the config `{communication_language}`
-- No human interaction: do not ask questions or wait for approval in this step.
-- Review subagents get no prior session context.
-- All review subagents must run at the same model capability as the current session.
-
-## INSTRUCTIONS
-
-Change `{spec_file}` status to `in-review` in the frontmatter before continuing.
-
-### Construct Diff
-
-Read `{baseline_revision}` from `{spec_file}` frontmatter. If `{baseline_revision}` is missing or `NO_VCS`, use best effort to determine what changed. Otherwise, construct `{diff_output}` covering all changes — tracked and untracked — since `{baseline_revision}`.
-
-Do NOT `git add` anything — this is read-only inspection.
-
-### Review
-
-Launch two subagents without prior session context.
-
-- **Blind hunter** — receives inline `{diff_output}` only. No spec, no context docs, no project access. Invoke via the `bmad-review-adversarial-general` skill.
-- **Edge case hunter** — receives `{diff_output}` and read access to the project. Invoke via the `bmad-review-edge-case-hunter` skill.
-
-### Classify
-
-1. Deduplicate all review findings.
-2. Classify each finding. The first three categories are **this story's problem** — caused or exposed by the current change. The last two are **not this story's problem**.
-   - **intent_gap** — caused by the change; cannot be resolved from the spec because the captured intent is incomplete. Do not infer intent unless there is exactly one possible reading.
-   - **bad_spec** — caused by the change, including direct deviations from spec. The spec should have been clear enough to prevent it. When in doubt between bad_spec and patch, prefer bad_spec — a spec-level fix is more likely to produce coherent code.
-   - **patch** — caused by the change; trivially fixable without human input. Just part of the diff.
-   - **defer** — pre-existing issue not caused by this story, surfaced incidentally by the review. Collect for later focused attention.
-   - **reject** — noise. Drop silently. When unsure between defer and reject, prefer reject — only defer findings you are confident are real.
-3. Process findings in cascading order. If intent_gap exists, lower findings are moot; follow the intent_gap branch below. If bad_spec exists, lower findings are moot since code will be re-derived. If neither exists, process patch and defer normally. Before each bad_spec loopback, read `{spec_file}` frontmatter `review_loop_iteration` (missing means `0`), increment it by 1, and write it back. If it exceeds 5, HALT with status `blocked` and blocking condition `review repair loop exceeded 5 iterations`.
-   - **intent_gap** — Root cause is inside `<intent-contract>`. Revert code changes. HALT with status `blocked`, blocking condition `intent gap in intent contract`, and include the intent-gap findings.
-   - **bad_spec** — Root cause is outside `<intent-contract>`. Do not modify content inside `<intent-contract>`. Before reverting code: extract KEEP instructions for positive preservation (what worked well and must survive re-derivation). Revert code changes. Read the `## Spec Change Log` in `{spec_file}` and strictly respect all logged constraints when amending the sections outside `<intent-contract>` that contain the root cause. Append a new change-log entry recording: the triggering finding, what was amended, the known-bad state avoided, and the KEEP instructions. Read fully and follow `./step-03-implement.md` to re-derive the code, then this step will run again.
-   - **patch** — Auto-fix. These are the only findings that survive loopbacks.
-   - **defer** — Append one new entry to `{deferred_work_file}` using this format. Do not modify existing entries or look for duplicates.
-     ```markdown
-     - source_spec: `{spec_file}`
-       summary: <one sentence>
-       evidence: <why this is real>
-     ```
-   - **reject** — Drop silently.
-
-## Finalize
-
-Prepare `Auto Run Result` details:
-- Summary of implemented change
-- Files changed with one-line descriptions
-- Review findings breakdown: patches applied, items deferred, items rejected
-- Verification performed, including command outcomes or manual inspection notes
-- Any residual risks
-
-HALT with status `done`.

src/bmm-skills/4-implementation/bmad-quick-dev/SKILL.md

@@ -20,8 +20,6 @@ A specification is "Ready for Development" when:
 - **Logical**: Tasks ordered by dependency.
 - **Testable**: All ACs use Given/When/Then.
 - **Complete**: No placeholders or TBDs.
-- **Sufficient**: No known requirement, acceptance, dependency, or implementation gaps remain unresolved.
-- **Coherent**: No unresolved ambiguities or internal contradictions.
 
 ## SCOPE STANDARD
 

src/bmm-skills/4-implementation/bmad-quick-dev/customize.toml

@@ -1,33 +1,41 @@
 # DO NOT EDIT -- overwritten on every update.
 #
-# Default customization values for bmad-quick-dev.
-# Override in _bmad/custom/bmad-quick-dev.toml or
-# _bmad/custom/bmad-quick-dev.user.toml.
-#
-# Merge rules:
-# - Strings replace the default.
-# - Lists append to the default list.
-# - Tables merge key by key.
+# Workflow customization surface for bmad-quick-dev. Mirrors the
+# agent customization shape under the [workflow] namespace.
 
 [workflow]
 
-# Extra instructions to run before config is loaded and before the user is greeted.
+# --- Configurable below. Overrides merge per BMad structural rules: ---
+#   scalars: override wins • arrays (persistent_facts, activation_steps_*): append
+#   arrays-of-tables with `code`/`id`: replace matching items, append new ones.
+
+# Steps to run before the standard activation (config load, greet).
+# Overrides append. Use for pre-flight loads, compliance checks, etc.
 
 activation_steps_prepend = []
 
-# Extra instructions to run after the greeting and before step 01.
+# Steps to run after greet but before the workflow begins.
+# Overrides append. Use for context-heavy setup that should happen
+# once the user has been acknowledged.
 
 activation_steps_append = []
 
-# Facts kept in context for the whole run.
-# Entries are literal text or file references prefixed with "file:".
-# File entries may use globs and are loaded during activation.
+# Persistent facts the workflow keeps in mind for the whole run
+# (standards, compliance constraints, stylistic guardrails).
+# Distinct from the runtime memory sidecar — these are static context
+# loaded on activation. Overrides append.
+#
+# Each entry is either:
+#   - a literal sentence, e.g. "All stories must include testable acceptance criteria."
+#   - a file reference prefixed with `file:`, e.g. "file:{project-root}/docs/standards.md"
+#     (glob patterns are supported; the file's contents are loaded and treated as facts).
 
 persistent_facts = [
   "file:{project-root}/**/project-context.md",
 ]
 
-# Instruction run after Quick Dev completes.
-# Empty means no extra completion behavior.
+# Scalar: executed when the workflow reaches its final step,
+# after implementation is complete and explanations are provided. Override wins.
+# Leave empty for no custom post-completion behavior.
 
 on_complete = ""

src/bmm-skills/4-implementation/bmad-quick-dev/spec-template.md

@@ -3,7 +3,6 @@ title: '{title}'
 type: 'feature' # feature | bugfix | refactor | chore
 created: '{date}'
 status: 'draft' # draft | ready-for-dev | in-progress | in-review | done
-review_loop_iteration: 0 # incremented by step-04 before each review loopback
 context: [] # optional: `{project-root}/`-prefixed paths to project-wide standards/docs the implementation agent should load. Keep short — only what isn't already distilled into the spec body.
 ---
 

src/bmm-skills/4-implementation/bmad-quick-dev/step-01-clarify-and-route.md

@@ -59,8 +59,8 @@ If the spec is an epic story and `{sprint_status}` exists: find the `development
         - **If missing, empty, or invalid:** continue to step 3.
 
      3. **Compile epic context.** Produce `{implementation_artifacts}/epic-<N>-context.md` by following `./compile-epic-context.md`, in order of preference:
-        - **Preferred — subagent:** spawn a subagent with `./compile-epic-context.md` as its prompt. Pass it the epic number, the epics file path, the `{planning_artifacts}` directory, and the output path `{implementation_artifacts}/epic-<N>-context.md`.
-        - **Fallback — inline** (for runtimes without subagent support, e.g. Copilot, Codex, local Ollama, older Claude): if your runtime cannot spawn subagents, or the spawn fails/times out, read `./compile-epic-context.md` yourself and follow its instructions to produce the same output file.
+        - **Preferred — sub-agent:** spawn a sub-agent with `./compile-epic-context.md` as its prompt. Pass it the epic number, the epics file path, the `{planning_artifacts}` directory, and the output path `{implementation_artifacts}/epic-<N>-context.md`.
+        - **Fallback — inline** (for runtimes without sub-agent support, e.g. Copilot, Codex, local Ollama, older Claude): if your runtime cannot spawn sub-agents, or the spawn fails/times out, read `./compile-epic-context.md` yourself and follow its instructions to produce the same output file.
 
      4. **Verify.** After compilation, verify the output file exists, is non-empty, and starts with `# Epic <N> Context:`. If valid, load it. If verification fails, HALT and report the failure.
 
@@ -82,12 +82,7 @@ If the spec is an epic story and `{sprint_status}` exists: find the `development
    - Present detected distinct goals as a bullet list.
    - Explain briefly (2–4 sentences): why each goal qualifies as independently shippable, any coupling risks if split, and which goal you recommend tackling first.
    - HALT and ask human: `[S] Split — pick first goal, defer the rest` | `[K] Keep all goals — accept the risks`
-   - On **S**: For each deferred goal, append one new entry to `{deferred_work_file}` using this format. Do not modify existing entries or look for duplicates. Narrow scope to the first-mentioned goal. Continue routing.
-     ```markdown
-     - source_spec: none
-       summary: <one sentence naming the deferred goal>
-       evidence: <why this was split from the current intent>
-     ```
+   - On **S**: Append deferred goals to `{deferred_work_file}`. Narrow scope to the first-mentioned goal. Continue routing.
    - On **K**: Proceed as-is.
 5. Route — choose exactly one:
 

src/bmm-skills/4-implementation/bmad-quick-dev/step-02-plan.md

@@ -12,19 +12,14 @@ deferred_work_file: '{implementation_artifacts}/deferred-work.md'
 ## INSTRUCTIONS
 
 1. Draft resume check. If `{spec_file}` exists with `status: draft`, read it and capture the verbatim `<frozen-after-approval>...</frozen-after-approval>` block as `preserved_intent`. Otherwise `preserved_intent` is empty.
-2. Investigate codebase. _Isolate deep exploration in subagents/tasks where available. To prevent context snowballing, instruct subagents to give you distilled summaries only._
+2. Investigate codebase. _Isolate deep exploration in sub-agents/tasks where available. To prevent context snowballing, instruct subagents to give you distilled summaries only._
 3. Read `./spec-template.md` fully. Fill it out based on the intent and investigation. If `{preserved_intent}` is non-empty, substitute it for the `<frozen-after-approval>` block in your filled spec before writing. Write the result to `{spec_file}`.
 4. Self-review against READY FOR DEVELOPMENT standard.
 5. If intent gaps exist, do not fantasize, do not leave open questions, HALT and ask the human.
 6. Token count check (see SCOPE STANDARD). If spec exceeds 1600 tokens:
    - Show user the token count.
    - HALT and ask human: `[S] Split — carve off secondary goals` | `[K] Keep full spec — accept the risks`
-   - On **S**: Propose the split — name each secondary goal. For each deferred goal, append one new entry to `{deferred_work_file}` using this format. Do not modify existing entries or look for duplicates. Rewrite the current spec to cover only the main goal — do not surgically carve sections out; regenerate the spec for the narrowed scope. Continue to checkpoint.
-     ```markdown
-     - source_spec: `{spec_file}`
-       summary: <one sentence naming the deferred goal>
-       evidence: <why this was split from the current spec>
-     ```
+   - On **S**: Propose the split — name each secondary goal. Append deferred goals to `{deferred_work_file}`. Rewrite the current spec to cover only the main goal — do not surgically carve sections out; regenerate the spec for the narrowed scope. Continue to checkpoint.
    - On **K**: Continue to checkpoint with full spec.
 
 ### CHECKPOINT 1

src/bmm-skills/4-implementation/bmad-quick-dev/step-03-implement.md

@@ -26,15 +26,15 @@ Change `{spec_file}` status to `in-progress` in the frontmatter before starting
 
 Follow `./sync-sprint-status.md` with `{target_status}` = `in-progress`.
 
-If `{spec_file}` has a non-empty `context:` list in its frontmatter, load those files before implementation begins. When handing to a subagent, include them in the subagent prompt so it has access to the referenced context.
+If `{spec_file}` has a non-empty `context:` list in its frontmatter, load those files before implementation begins. When handing to a sub-agent, include them in the sub-agent prompt so it has access to the referenced context.
 
-Hand `{spec_file}` to a subagent/task and let it implement. If no subagents are available, implement directly.
+Hand `{spec_file}` to a sub-agent/task and let it implement. If no sub-agents are available, implement directly.
 
 **Path formatting rule:** Any markdown links written into `{spec_file}` must use paths relative to `{spec_file}`'s directory so they are clickable in VS Code. Any file paths displayed in terminal/conversation output must use CWD-relative format with `:line` notation (e.g., `src/path/file.ts:42`) for terminal clickability. No leading `/` in either case.
 
-### Tasks & Acceptance Verification
+### Self-Check
 
-Before leaving this step, verify every task in the `## Tasks & Acceptance` section of `{spec_file}` is complete and every acceptance criterion is satisfied. Mark each finished task `[x]`. If any task is not done or any acceptance criterion is not satisfied, finish the missing work before proceeding.
+Before leaving this step, verify every task in the `## Tasks & Acceptance` section of `{spec_file}` is complete. Mark each finished task `[x]`. If any task is not done, finish it before proceeding.
 
 ## NEXT
 

src/bmm-skills/4-implementation/bmad-quick-dev/step-04-review.md

@@ -1,5 +1,6 @@
 ---
 deferred_work_file: '{implementation_artifacts}/deferred-work.md'
+specLoopIteration: 1
 ---
 
 # Step 4: Review
@@ -22,10 +23,11 @@ Do NOT `git add` anything — this is read-only inspection.
 
 ### Review
 
-Launch two subagents without conversation context. If no subagents are available, generate two review prompt files in `{implementation_artifacts}` — one per reviewer role below — and HALT. Ask the human to run each in a separate session (ideally a different LLM) and paste back the findings.
+Launch three subagents without conversation context. If no sub-agents are available, generate three review prompt files in `{implementation_artifacts}` — one per reviewer role below — and HALT. Ask the human to run each in a separate session (ideally a different LLM) and paste back the findings.
 
 - **Blind hunter** — receives inline `{diff_output}` only. No spec, no context docs, no project access. Invoke via the `bmad-review-adversarial-general` skill.
 - **Edge case hunter** — receives `{diff_output}` and read access to the project. Invoke via the `bmad-review-edge-case-hunter` skill.
+- **Acceptance auditor** — receives `{diff_output}`, `{spec_file}`, and read access to the project. Must also read the docs listed in `{spec_file}` frontmatter `context`. Checks for violations of acceptance criteria, rules, and principles from the spec and context docs.
 
 ### Classify
 
@@ -36,16 +38,11 @@ Launch two subagents without conversation context. If no subagents are available
    - **patch** — caused by the change; trivially fixable without human input. Just part of the diff.
    - **defer** — pre-existing issue not caused by this story, surfaced incidentally by the review. Collect for later focused attention.
    - **reject** — noise. Drop silently. When unsure between defer and reject, prefer reject — only defer findings you are confident are real.
-3. Process findings in cascading order. If intent_gap or bad_spec findings exist, they trigger a loopback — lower findings are moot since code will be re-derived. If neither exists, process patch and defer normally. Before each loopback, read `{spec_file}` frontmatter `review_loop_iteration` (missing means `0`), increment it by 1, and write it back. If it exceeds 5, HALT and escalate to the human.
+3. Process findings in cascading order. If intent_gap or bad_spec findings exist, they trigger a loopback — lower findings are moot since code will be re-derived. If neither exists, process patch and defer normally. Increment `{specLoopIteration}` on each loopback. If it exceeds 5, HALT and escalate to the human.
    - **intent_gap** — Root cause is inside `<frozen-after-approval>`. Revert code changes. Loop back to the human to resolve. Once resolved, read fully and follow `./step-02-plan.md` to re-run steps 2–4.
    - **bad_spec** — Root cause is outside `<frozen-after-approval>`. Before reverting code: extract KEEP instructions for positive preservation (what worked well and must survive re-derivation). Revert code changes. Read the `## Spec Change Log` in `{spec_file}` and strictly respect all logged constraints when amending the non-frozen sections that contain the root cause. Append a new change-log entry recording: the triggering finding, what was amended, the known-bad state avoided, and the KEEP instructions. Read fully and follow `./step-03-implement.md` to re-derive the code, then this step will run again.
    - **patch** — Auto-fix. These are the only findings that survive loopbacks.
-   - **defer** — Append one new entry to `{deferred_work_file}` using this format. Do not modify existing entries or look for duplicates.
-     ```markdown
-     - source_spec: `{spec_file}`
-       summary: <one sentence>
-       evidence: <why this is real>
-     ```
+   - **defer** — Append to `{deferred_work_file}`.
    - **reject** — Drop silently.
 
 ## NEXT

src/bmm-skills/4-implementation/bmad-quick-dev/step-oneshot.md

@@ -19,19 +19,14 @@ Implement the clarified intent directly.
 
 ### Review
 
-Invoke the `bmad-review-adversarial-general` skill in a subagent with the changed files. The subagent gets NO conversation context — to avoid anchoring bias. Launch at the same model capability as the current session. If no subagents are available, write the changed files to a review prompt file in `{implementation_artifacts}` and HALT. Ask the human to run the review in a separate session and paste back the findings.
+Invoke the `bmad-review-adversarial-general` skill in a subagent with the changed files. The subagent gets NO conversation context — to avoid anchoring bias. Launch at the same model capability as the current session. If no sub-agents are available, write the changed files to a review prompt file in `{implementation_artifacts}` and HALT. Ask the human to run the review in a separate session and paste back the findings.
 
 ### Classify
 
 Deduplicate all review findings. Three categories only:
 
 - **patch** — trivially fixable. Auto-fix immediately.
-- **defer** — pre-existing issue not caused by this change. Append one new entry to `{deferred_work_file}` using this format. Do not modify existing entries or look for duplicates.
-  ```markdown
-  - source_spec: `{spec_file}`
-    summary: <one sentence>
-    evidence: <why this is real>
-  ```
+- **defer** — pre-existing issue not caused by this change. Append to `{deferred_work_file}`.
 - **reject** — noise. Drop silently.
 
 If a finding is caused by this change but too significant for a trivial patch, HALT and present it to the human for decision before proceeding.

src/core-skills/bmad-brainstorming/SKILL.md

@@ -18,7 +18,7 @@ The session runs in one of three stances, chosen by the user — set explicitly
 
 ## On Activation
 
-1. Resolve customization: `uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, use a subagent to read `{skill-root}/customize.toml` directly with defaults.
+1. Resolve customization: `python3 {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, use a subagent to read `{skill-root}/customize.toml` directly with defaults.
 2. Run each `{workflow.activation_steps_prepend}` entry. Treat each `{workflow.persistent_facts}` entry as foundational context (`file:`-prefixed entries are paths/globs under `{project-root}` — load their contents; others are facts verbatim).
 3. Load `{project-root}/_bmad/core/config.yaml` (and `config.user.yaml` if present); resolve `{user_name}`, `{communication_language}`, `{document_output_language}`, `{output_folder}`, `{project_name}`, `{date}`. Missing → neutral defaults; never block.
 4. **If launched headless** (a machine signal, not a human asking for output — `references/headless.md` lists them): load `references/headless.md` and follow it for the whole run. It is the *only* context where you generate ideas yourself; never load it otherwise.
@@ -34,19 +34,21 @@ These fight your defaults, in every mode; hold them deliberately. The stance you
 - **Keep shifting the creative domain** — every 5–10 turns (or ~10 ideas when you're generating), usually by moving to the next technique.
 - **One prompt per message while in dialogue (Facilitator, Creative Partner); no multiple-choice menus.** Don't stack questions into a wall or hand a menu that invites lazy picking — both pull the user out of generating. The only exceptions are the two up-front *process* choices (stance, and the technique flow): *how* to run is theirs to pick; *what* to ideate never is.
 
-**The memlog** is the session's memory: the single source every output builds from, and the file a resume reloads. Whatever isn't in it is gone. Log every idea, decision, question, and bit of user direction — anything you'd regret losing if the window closed — one line each, the gist in the user's meaning, in time order; never edit or reorder. Skip your prompts and small talk. All writes to memlog are atomic and use the script `memlog.py` invoked as follows:
+**The memlog** is the session's memory: the single source every output builds from, and the file a resume reloads. Whatever isn't in it is gone. Log every idea, decision, question, and bit of user direction — anything you'd regret losing if the window closed — one line each, the gist in the user's meaning, in time order; never edit or reorder. Skip your prompts and small talk. All writes go through `scripts/memlog.py` (atomic; don't read it back mid-session — resume is the one exception):
 
-- `uv run {project-root}/_bmad/scripts/memlog.py init --workspace {doc_workspace} --field topic="<topic>" --field goal="<goal>" --field mode="<facilitator|partner|autonomous>"` — create it once topic, goal, and stance are known.
-- `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type <kind> --text "<one-line gist>"` — log one entry. `--type` ∈ `idea`/`insight`/`question`/`decision`/`direction`/`technique` (a switch: `--text "started <name>"`); omit for a plain note. Add `--by user`/`--by coach` to mark authorship — **required in Creative Partner mode** (renders `(idea by user)`); skip it otherwise.
-- `uv run {project-root}/_bmad/scripts/memlog.py set --workspace {doc_workspace} --key status --value complete` — flip status at wrap-up.
+- `memlog.py init --workspace {doc_workspace} --field topic="<topic>" --field goal="<goal>" --field mode="<facilitator|partner|autonomous>"` — create it once topic, goal, and stance are known.
+- `memlog.py append --workspace {doc_workspace} --type <kind> --text "<one-line gist>"` — log one entry. `--type` ∈ `idea`/`insight`/`question`/`decision`/`direction`/`technique` (a switch: `--text "started <name>"`); omit for a plain note. Add `--by user`/`--by coach` to mark authorship — **required in Creative Partner mode** (renders `(idea by user)`); skip it otherwise.
+- `memlog.py set --workspace {doc_workspace} --key status --value complete` — flip status at wrap-up.
+
+(Each is `python3 {skill-root}/scripts/memlog.py …`.)
 
 ## Run a Session
 
-Open with one compound question what are we brainstorming, and what's the goal or why behind it (along with asking if there are any inputs or special requests). The why shapes technique choice and synthesis (*kids' iPhone apps to build with your own kids* vs. *to win market share* point different ways). If the kickoff already made both clear, skip the question and confirm; read anything they point you to. Derive a kebab-case `{topic_slug}` and bind `{doc_workspace} = {workflow.output_dir}/{workflow.output_folder_name}/`.
+Open with one compound question — **what are we brainstorming, and what's the goal or why behind it?** (plus any inputs or special requests). The why shapes technique choice and synthesis (*kids' iPhone apps to build with your own kids* vs. *to win market share* point different ways). If the kickoff already made both clear, skip the question and confirm; read anything they point you to. Derive a kebab-case `{topic_slug}` and bind `{doc_workspace} = {workflow.output_dir}/{workflow.output_folder_name}/`.
 
 Now set the **stance** and the **technique batch** in one step — the composer page does both, so make it the default.
 
-**The composer page (primary).** The file is `{skill-root}/assets/brain-selector.html`. With a customized catalog (overridden `{workflow.brain_methods}` or any `{workflow.additional_techniques}`), regenerate it first: `uv run {skill-root}/scripts/brain.py --file {workflow.brain_methods} [--extra {doc_workspace}/extra-techniques.json] html --out {doc_workspace}/brain-selector.html` (pass `--extra`, a JSON list of `{category, technique_name, description}`, when there are additional techniques; the file is then `{doc_workspace}/brain-selector.html`). Try to open it (`open` / `xdg-open` / `start`), then say, in one message: *"It should open in your browser — compose your session, click **Copy prompt**, and paste the result back. If it didn't open, open `<path>` yourself, or say 'let's do it in chat'."* You can't see their browser, so never claim it opened.
+**The composer page (primary).** The file is `{skill-root}/assets/brain-selector.html`. With a customized catalog (overridden `{workflow.brain_methods}` or any `{workflow.additional_techniques}`), regenerate it first: `python3 {skill-root}/scripts/brain.py --file {workflow.brain_methods} [--extra {doc_workspace}/extra-techniques.json] html --out {doc_workspace}/brain-selector.html` (pass `--extra`, a JSON list of `{category, technique_name, description}`, when there are additional techniques; the file is then `{doc_workspace}/brain-selector.html`). Try to open it (`open` / `xdg-open` / `start`), then say, in one message: *"It should open in your browser — compose your session, click **Copy prompt**, and paste the result back. If it didn't open, open `<path>` yourself, or say 'let's do it in chat'."* You can't see their browser, so never claim it opened.
 
 Read the pasted block: the **`Facilitation mode:`** line → the stance; the **listed techniques** (full category/name/description, some tagged `(random pick)`) → run them as given, no `list`/`show` needed; **`invent N`** / **`you choose N`** → see `## Choosing Techniques`.
 
@@ -61,7 +63,7 @@ For **Facilitator** and **Creative Partner**. (In **Ideate for me** you pick and
 Most sessions arrive with a batch already composed on the page — run it as given (each technique's full text is in the paste; no `list`/`show` needed). Two parts of a paste delegate back to you:
 
 - **`invent N`** (Inventive Flow) — invent N brand-new techniques on the fly. A line may scope an invention (`invent 1 new technique in the spirit of <category>`, from the page's per-category invent card) — when it does, honor that category's spirit. Announce the order, log each one's name + description, and offer to save a keeper to `{workflow.additional_techniques}` at wrap-up.
-- **`you choose N`** (Facilitator Chosen) — pick N techniques fitting the goal, `{workflow.favorite_techniques}` first; confirm exact names with a scoped `uv run {skill-root}/scripts/brain.py --file {workflow.brain_methods} list --category <cat>`. Never pull the library whole into context.
+- **`you choose N`** (Facilitator Chosen) — pick N techniques fitting the goal, `{workflow.favorite_techniques}` first; confirm exact names with a scoped `python3 {skill-root}/scripts/brain.py --file {workflow.brain_methods} list --category <cat>`. Never pull the library whole into context.
 
 If they didn't use the page, load `references/in-chat-techniques.md` and pick the batch in chat (**3–4 is the sweet spot**).
 

src/core-skills/bmad-brainstorming/references/converge.md

@@ -17,7 +17,7 @@ Pick by what the decision needs:
 - **PMI (Plus / Minus / Interesting)** — when one strong candidate needs pressure-testing before commitment: list its pluses, minuses, and the merely-interesting, then judge.
 - **MoSCoW** — when scoping a build: sort into Must / Should / Could / Won't-this-time.
 
-Log the surviving directions and the reasoning with `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type decision --text "<one-line gist>"` (use `--by` in Creative Partner mode). Two or three convergence moves chained is fine (e.g. cluster → score the clusters); more than that is usually over-processing.
+Log the surviving directions and the reasoning with `python3 {skill-root}/scripts/memlog.py append --type decision --text "<one-line gist>"` (use `--by` in Creative Partner mode). Two or three convergence moves chained is fine (e.g. cluster → score the clusters); more than that is usually over-processing.
 
 ## Then finalize
 

src/core-skills/bmad-brainstorming/references/finalize.md

@@ -9,7 +9,7 @@ In Facilitator mode this is the one place your own creative contribution is welc
 1. **Hand them the mirror first.** Reflect a vivid sampling of *their* ideas back — deliberately include the odd, random, or buried ones from earlier, not just the recent obvious ones (in Creative Partner mode the `(... by user)` tags tell you which were theirs). Ask what they see now: conclusions, synergies, themes, the few that actually matter. Let them connect first; their own pattern-recognition is the point.
 2. **Then add the connections they would miss.** Lean in creatively — not new raw ideas, but the non-obvious links: this idea from technique one quietly solves that tension from technique four; these three are one idea wearing three hats; this wildcard is the real breakthrough.
 
-Record the insights and chosen directions with `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type insight --text "<insights + chosen directions>"`. **Then run `uv run {project-root}/_bmad/scripts/memlog.py set --workspace {doc_workspace} --key status --value complete`** — the session is done and must stop being offered for resume. Do this even if the user declines every artifact below.
+Record the insights and chosen directions with `memlog.py append --type insight`. **Then run `python3 {skill-root}/scripts/memlog.py set --workspace {doc_workspace} --key status --value complete`** — the session is done and must stop being offered for resume. Do this even if the user declines every artifact below.
 
 ## Artifacts
 

src/core-skills/bmad-brainstorming/references/headless.md

@@ -14,7 +14,7 @@ When in doubt, you are interactive — a present human asking you to "brainstorm
 
 ## The inversion
 
-There is no user to draw ideas out of, so you become the brainstormer. Run a real divergent session against the supplied topic: discover techniques with `uv run {skill-root}/scripts/brain.py --file {workflow.brain_methods} list --all` (the whole catalog is fine here — you are generating, not pacing a user; add `show "<name>"` for a technique's full method on demand), plus any `{workflow.additional_techniques}`, preferring `{workflow.favorite_techniques}` where they fit; work them, and **shift the creative domain every ~10 ideas** exactly as the interactive Stance demands — technical, then experiential, then business, then failure modes, then wildcards. Push past the obvious; the same quantity ambition (aim past 100) and anti-clustering discipline apply. The only thing that changes is that the ideas are now yours to generate. This relaxation is scoped entirely to this file — it never applies to interactive sessions.
+There is no user to draw ideas out of, so you become the brainstormer. Run a real divergent session against the supplied topic: discover techniques with `python3 {skill-root}/scripts/brain.py --file {workflow.brain_methods} list --all` (the whole catalog is fine here — you are generating, not pacing a user; add `show "<name>"` for a technique's full method on demand), plus any `{workflow.additional_techniques}`, preferring `{workflow.favorite_techniques}` where they fit; work them, and **shift the creative domain every ~10 ideas** exactly as the interactive Stance demands — technical, then experiential, then business, then failure modes, then wildcards. Push past the obvious; the same quantity ambition (aim past 100) and anti-clustering discipline apply. The only thing that changes is that the ideas are now yours to generate. This relaxation is scoped entirely to this file — it never applies to interactive sessions.
 
 ## Inputs the caller is expected to provide
 
@@ -29,9 +29,9 @@ Free-form structured payload in the first message; provide what applies:
 
 ## Run
 
-1. Bind `{doc_workspace}` and create the memlog with `uv run {project-root}/_bmad/scripts/memlog.py init --workspace {doc_workspace} --field topic="<topic>" [--field goal="<goal>"]`. It remains the canonical source every artifact derives from.
-2. Run the divergent session per **The inversion**, capturing each idea with `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type idea --text "<idea>"` as it lands, and marking each technique switch with `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type technique --text "started <name>"`.
-3. Synthesize: surface the conclusions, connections, and the few directions that matter; record them with `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {doc_workspace} --type insight --text "<insights>"`, then run `uv run {project-root}/_bmad/scripts/memlog.py set --workspace {doc_workspace} --key status --value complete`.
+1. Bind `{doc_workspace}` and create the memlog with `python3 {skill-root}/scripts/memlog.py init --workspace {doc_workspace} --field topic="<topic>" [--field goal="<goal>"]`. It remains the canonical source every artifact derives from.
+2. Run the divergent session per **The inversion**, capturing each idea with `memlog.py append --workspace {doc_workspace} --type idea --text "<idea>"` as it lands, and marking each technique switch with `memlog.py append --type technique --text "started <name>"`.
+3. Synthesize: surface the conclusions, connections, and the few directions that matter; record them with `memlog.py append --type insight`, then run `memlog.py set --workspace {doc_workspace} --key status --value complete`.
 4. Produce the requested artifacts from the log — `brainstorm.html` (the imaginative, self-contained, no-template report) and/or the succinct `brainstorm-intent.md` — the same artifacts `references/finalize.md` describes, delegating each to a subagent that reads the log as its sole source. (Headless produces the `artifacts` payload directly; it does not ask, unlike the interactive opt-in.)
 5. Execute each entry in `{workflow.external_handoffs}` (capture returned URLs/IDs into the JSON `external_handoffs` array; skip and flag unavailable tools — local files always exist). Then run `{workflow.on_complete}` if non-empty.
 

src/core-skills/bmad-brainstorming/references/in-chat-techniques.md

@@ -7,7 +7,7 @@ Loaded only when the user won't use the composer page (no browser, headless, or
 - **Category** — the user names 1–n categories; `random --category` draws the batch from them. No listing needed.
 - **Inventive Flow** — invent at least 3 techniques, announce the order before the first, touch no script. Log each one's name + description so you can offer to save a keeper to `{workflow.additional_techniques}` (via `bmad-customize`) at wrap-up.
 
-The library is large — never pull it whole into context. The only way in is the helper, always passing `--file {workflow.brain_methods}`. Subcommands of `uv run {skill-root}/scripts/brain.py --file {workflow.brain_methods}`:
+The library is large — never pull it whole into context. The only way in is the helper, always passing `--file {workflow.brain_methods}`. Subcommands of `python3 {skill-root}/scripts/brain.py --file {workflow.brain_methods}`:
 
 - `categories` — names + counts; the cheap survey map.
 - `list --category X [--category Y]` — the index (name + gist) for those categories. Bare `list` is refused by the script.

src/core-skills/bmad-brainstorming/references/mode-autonomous.md

@@ -5,6 +5,6 @@ The user handed you the topic and wants to see what you come up with on your own
 - **Run a real divergent session yourself.** Pick and run techniques on your own (use `brain.py` as in `## Choosing Techniques`, but *you* choose — no menu for the user), capturing each idea to the memlog with `--type idea --by coach`, marking each technique switch with a `technique` entry, shifting the creative domain every ~10 ideas, aiming past 100. Push past the obvious.
 - **Don't pepper the user with questions** — this is your run. One quick confirm of topic and goal up front is plenty.
 - **When it's mined out, synthesize and produce the keepsake.** Go to `## Wrap-Up` (`references/finalize.md`): record the insights, mark the memlog complete, and **auto-generate the imaginative HTML keepsake — don't ask first; the keepsake is the result you promised to show them.** Offer the other artifacts (intent doc, etc.) after.
-- **Then, because a human is here, offer to keep going together.** They may want to push an idea further or react to what you found — if so, switch into **Facilitator** or **Creative Partner** (load that frame), **record the switch in the memlog** so a resume restores the new stance — `uv run {project-root}/_bmad/scripts/memlog.py set --workspace {doc_workspace} --key mode --value <facilitator|partner>` — and continue from the same memlog.
+- **Then, because a human is here, offer to keep going together.** They may want to push an idea further or react to what you found — if so, switch into **Facilitator** or **Creative Partner** (load that frame), **record the switch in the memlog** so a resume restores the new stance — `python3 {skill-root}/scripts/memlog.py set --workspace {doc_workspace} --key mode --value <facilitator|partner>` — and continue from the same memlog.
 
 This is the interactive sibling of headless mode (`references/headless.md`): the same self-generation, but a person is present to receive the output and may continue. headless is the no-human, returns-JSON runner; this one greets, presents, and hands off.

src/core-skills/bmad-brainstorming/scripts/memlog.py

@@ -0,0 +1,202 @@
+#!/usr/bin/env python3
+# /// script
+# requires-python = ">=3.10"
+# ///
+"""memlog — an append-only memory log: LLM-optimal working memory for a skill.
+
+A memlog is the dense, chronological record of everything that mattered in a piece of
+work — every item the user generated or accepted — kept minimal like human memory: only
+what's important, never bloated. It persists ACROSS sessions, so a fresh session can
+load it and continue. It is NOT a deliverable; downstream artifacts (a brief, a PRD, a
+deck, a report) are *derived* from it on demand. The host skill supplies the vocabulary
+by how it calls `append` — the tool stays neutral.
+
+It is a FLAT log: there are no sections or grouping. Every entry is one line, recorded
+at the END in the order it happened. The chronology itself is the structure — an event
+like "started technique X" is just another entry, same as an idea or an insight.
+
+Two invariants make it trustworthy:
+
+  1. Append-only, chronological. Entries land at the end, in the order they happen.
+     Nothing is ever inserted backward, reordered, or grouped.
+  2. Write-only / blind. Every command is an atomic, context-free write and echoes the
+     new state as JSON, so the caller never re-reads the file mid-session. The one time
+     the file is read is on resume — and the caller reads it itself, not via this script.
+
+The file shape (.memlog.md):
+
+    ---
+    topic: Onboarding flow for a budgeting app
+    goal: lift week-1 retention
+    status: active
+    updated: 2026-05-30T14:22
+    ---
+
+    - (note) user picked techniques: SCAMPER, then Six Thinking Hats
+    - (technique) started SCAMPER
+    - (idea) skip the signup wall: let people try with sample data first
+    - (idea) auto-import one bank account so the first screen shows real numbers
+    - (question) is open-banking consent too heavy for step one?
+    - (technique) started Six Thinking Hats
+    - (idea) black-hat: imported transactions look scary before they're categorized
+    - (insight) the "scary numbers" risk and the "real numbers" idea are one lever: show real data, pre-categorized
+    - (direction) user wants to optimize for the anxious first-timer, not the power user
+    - (decision) lead with one pre-categorized account; defer multi-account import
+
+Each entry may carry an optional `--type` — what KIND it is (idea, insight, question,
+decision, technique, …) — and an optional `--by` naming who it came from (e.g. `user`,
+`coach`), for sessions where authorship matters. Both render into one short inline tag:
+`(idea)`, `(idea by user)`, `(by coach)`. Omit them for a plain note. The host skill
+names the vocabulary; the script does not.
+
+Commands:
+  init   --workspace DIR [--field k=v ...]              create the memlog (errors if it exists)
+  append --workspace DIR --text STR [--type T] [--by W]  append one entry at the end
+  set    --workspace DIR --key K --value V              set/replace a frontmatter field
+
+The workspace is the run folder; the memlog is always {workspace}/.memlog.md.
+"""
+import argparse
+import json
+import os
+import sys
+from datetime import datetime
+from pathlib import Path
+
+MEMLOG = ".memlog.md"
+
+
+def now() -> str:
+    return datetime.now().strftime("%Y-%m-%dT%H:%M")
+
+
+def memlog_path(workspace: str) -> Path:
+    return Path(workspace) / MEMLOG
+
+
+def split(text: str) -> tuple[dict, str]:
+    """Return (frontmatter dict in source order, body str). Frontmatter is plain key: value.
+
+    The closing fence is the first line that is *exactly* `---`, so a `---` inside a
+    field value (topic/goal are free user text) never truncates the frontmatter.
+    """
+    lines = text.splitlines()
+    if not lines or lines[0] != "---":
+        raise ValueError(".memlog.md has no frontmatter")
+    end = next((i for i in range(1, len(lines)) if lines[i] == "---"), None)
+    if end is None:
+        raise ValueError(".memlog.md frontmatter is not terminated")
+    meta: dict[str, str] = {}
+    for line in lines[1:end]:
+        if ":" in line:
+            k, v = line.split(":", 1)
+            meta[k.strip()] = v.strip()
+    return meta, "\n".join(lines[end + 1:]).lstrip("\n")
+
+
+def render(meta: dict, body: str) -> str:
+    # Neutralize newlines in values so a multi-line field can't break the fence on re-read.
+    fm = "\n".join(f"{k}: {' '.join(str(v).splitlines())}" for k, v in meta.items())
+    return "---\n" + fm + "\n---\n\n" + body.rstrip("\n") + "\n"
+
+
+def touch(meta: dict) -> None:
+    """Stamp `updated` and keep it last so the field order stays predictable."""
+    meta.pop("updated", None)
+    meta["updated"] = now()
+
+
+def write_atomic(path: Path, text: str) -> None:
+    tmp = path.with_suffix(path.suffix + ".tmp")
+    tmp.write_text(text, encoding="utf-8")
+    os.replace(tmp, path)
+
+
+def entry_count(body: str) -> int:
+    return sum(1 for ln in body.splitlines() if ln.startswith("- "))
+
+
+def ack(path: Path, meta: dict, body: str) -> None:
+    """Echo new state so the caller never re-reads the file to know where it stands."""
+    print(json.dumps({
+        "ok": True,
+        "memlog": str(path),
+        "status": meta.get("status", ""),
+        "entries": entry_count(body),
+    }))
+
+
+def cmd_init(args) -> int:
+    path = memlog_path(args.workspace)
+    if path.exists():
+        print(f"error: {path} already exists; use append/set to update it", file=sys.stderr)
+        return 2
+    path.parent.mkdir(parents=True, exist_ok=True)
+    meta: dict[str, str] = {}
+    for pair in args.field or []:
+        if "=" not in pair:
+            print(f"error: --field expects key=value, got {pair!r}", file=sys.stderr)
+            return 2
+        k, v = pair.split("=", 1)
+        meta[k.strip()] = v.strip()
+    meta.setdefault("status", "active")
+    touch(meta)
+    write_atomic(path, render(meta, ""))
+    ack(path, meta, "")
+    return 0
+
+
+def cmd_append(args) -> int:
+    path = memlog_path(args.workspace)
+    meta, body = split(path.read_text(encoding="utf-8"))
+    text = " ".join(args.text.split())  # collapse newlines/runs → one-line entry, no prose bloat
+    label = args.type or ""
+    if args.by:
+        label = f"{label} by {args.by}".strip()  # attribution: "(idea by user)" / "(by coach)"
+    tag = f"({label}) " if label else ""
+    entry = f"- {tag}{text}"
+    body = (body.rstrip("\n") + "\n" + entry) if body.strip() else entry  # always at the end
+    touch(meta)
+    write_atomic(path, render(meta, body))
+    ack(path, meta, body)
+    return 0
+
+
+def cmd_set(args) -> int:
+    path = memlog_path(args.workspace)
+    meta, body = split(path.read_text(encoding="utf-8"))
+    meta[args.key] = args.value
+    touch(meta)
+    write_atomic(path, render(meta, body))
+    ack(path, meta, body)
+    return 0
+
+
+def main(argv: list[str] | None = None) -> int:
+    p = argparse.ArgumentParser(description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
+    sub = p.add_subparsers(dest="cmd", required=True)
+
+    pi = sub.add_parser("init", help="create the memlog")
+    pi.add_argument("--workspace", required=True)
+    pi.add_argument("--field", action="append", metavar="KEY=VALUE", help="frontmatter field (repeatable)")
+    pi.set_defaults(func=cmd_init)
+
+    pa = sub.add_parser("append", help="append one entry at the end")
+    pa.add_argument("--workspace", required=True)
+    pa.add_argument("--text", required=True)
+    pa.add_argument("--type", help="entry kind, rendered as an inline tag")
+    pa.add_argument("--by", help="who the entry came from (e.g. user, coach); rendered into the tag")
+    pa.set_defaults(func=cmd_append)
+
+    pset = sub.add_parser("set", help="set a frontmatter field")
+    pset.add_argument("--workspace", required=True)
+    pset.add_argument("--key", required=True)
+    pset.add_argument("--value", required=True)
+    pset.set_defaults(func=cmd_set)
+
+    args = p.parse_args(argv)
+    return args.func(args)
+
+
+if __name__ == "__main__":
+    sys.exit(main())

src/core-skills/bmad-brainstorming/scripts/tests/test_brain.py

@@ -207,11 +207,11 @@ def test_unknown_category_style_uses_fallback_glyph():
 
 def test_shipped_selector_is_in_sync_with_catalog():
     # foolproofing: if someone edits brain-methods.csv they must regenerate the page.
-    # Regenerate with: uv run brain.py html --out assets/brain-selector.html
+    # Regenerate with: python3 brain.py html --out assets/brain-selector.html
     asset = brain.DEFAULT_FILE.parent / "brain-selector.html"
     assert asset.is_file(), "missing assets/brain-selector.html — generate it"
     expected = brain.html_doc(brain.load(brain.DEFAULT_FILE))
     assert asset.read_text(encoding="utf-8") == expected, (
         "assets/brain-selector.html is stale; regenerate: "
-        "uv run brain.py html --out assets/brain-selector.html"
+        "python3 brain.py html --out assets/brain-selector.html"
     )

src/core-skills/bmad-brainstorming/scripts/tests/test_memlog.py

@@ -0,0 +1,265 @@
+# /// script
+# requires-python = ">=3.10"
+# dependencies = ["pytest>=8.0"]
+# ///
+"""Tests for memlog.py. Run: uv run --with pytest pytest scripts/tests/test_memlog.py
+
+The spine under test is the flat, append-only, chronological invariant: every entry is
+one line recorded at the end in the order it happened — no sections, no grouping.
+"""
+import json
+import sys
+from pathlib import Path
+
+import pytest
+
+sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
+import memlog  # noqa: E402
+
+MEMLOG = ".memlog.md"
+
+
+@pytest.fixture
+def ws(tmp_path):
+    return str(tmp_path)
+
+
+def read(ws):
+    return (Path(ws) / MEMLOG).read_text(encoding="utf-8")
+
+
+def body_of(ws):
+    return memlog.split(read(ws))[1]
+
+
+def entries(ws):
+    return [ln for ln in body_of(ws).splitlines() if ln.startswith("- ")]
+
+
+def init(ws, **fields):
+    fields = fields or {"topic": "Reinvent the lunchbox", "goal": "ideas for a pitch"}
+    argv = ["init", "--workspace", ws]
+    for k, v in fields.items():
+        argv += ["--field", f"{k}={v}"]
+    assert memlog.main(argv) == 0
+
+
+def append(ws, text, entry_type=None, by=None):
+    argv = ["append", "--workspace", ws, "--text", text]
+    if entry_type:
+        argv += ["--type", entry_type]
+    if by:
+        argv += ["--by", by]
+    assert memlog.main(argv) == 0
+
+
+# --- init ---------------------------------------------------------------
+
+def test_init_writes_frontmatter_fields(ws):
+    init(ws)
+    meta, body = memlog.split(read(ws))
+    assert meta["topic"] == "Reinvent the lunchbox"
+    assert meta["goal"] == "ideas for a pitch"
+    assert meta["status"] == "active"
+    assert "updated" in meta
+    assert body.strip() == ""
+
+
+def test_init_arbitrary_fields(ws):
+    init(ws, topic="T", audience="board")
+    meta, _ = memlog.split(read(ws))
+    assert meta["audience"] == "board"
+
+
+def test_init_refuses_overwrite(ws):
+    init(ws)
+    assert memlog.main(["init", "--workspace", ws, "--field", "topic=other"]) == 2
+
+
+def test_init_creates_missing_workspace(tmp_path):
+    nested = str(tmp_path / "a" / "b")
+    assert memlog.main(["init", "--workspace", nested, "--field", "topic=T"]) == 0
+    assert (Path(nested) / MEMLOG).is_file()
+
+
+def test_init_rejects_malformed_field(ws):
+    assert memlog.main(["init", "--workspace", ws, "--field", "noequals"]) == 2
+
+
+# --- append: flat chronological order is the whole point -----------------
+
+def test_append_lands_at_end_in_order(ws):
+    init(ws)
+    append(ws, "first")
+    append(ws, "second")
+    append(ws, "third")
+    assert entries(ws) == ["- first", "- second", "- third"]
+
+
+def test_no_sections_or_headings_ever(ws):
+    init(ws)
+    append(ws, "started foo", entry_type="technique")
+    append(ws, "an idea", entry_type="idea")
+    append(ws, "started bar", entry_type="technique")
+    assert "## " not in body_of(ws)  # the flat log never grows headings
+
+
+def test_type_renders_as_inline_tag(ws):
+    init(ws)
+    append(ws, "the earth revolves around the sun", entry_type="idea")
+    append(ws, "how do we handle stampede?", entry_type="question")
+    body = body_of(ws)
+    assert "- (idea) the earth revolves around the sun" in body
+    assert "- (question) how do we handle stampede?" in body
+
+
+def test_append_without_type_is_plain_note(ws):
+    init(ws)
+    append(ws, "bare entry")
+    assert entries(ws) == ["- bare entry"]
+
+
+def test_append_collapses_newlines_into_one_line(ws):
+    init(ws)
+    append(ws, "line one\nline two\n  spaced   out")
+    assert entries(ws) == ["- line one line two spaced out"]
+
+
+def test_revisited_technique_is_just_a_later_entry(ws):
+    # the user's model: switching techniques is an entry, not a section to return to
+    init(ws)
+    append(ws, "started SCAMPER", entry_type="technique")
+    append(ws, "magnetic latch", entry_type="idea")
+    append(ws, "started Six Hats", entry_type="technique")
+    append(ws, "stale data risk", entry_type="idea")
+    append(ws, "started SCAMPER", entry_type="technique")  # back to SCAMPER — just appended again
+    append(ws, "stackable tiers", entry_type="idea")
+    assert entries(ws) == [
+        "- (technique) started SCAMPER",
+        "- (idea) magnetic latch",
+        "- (technique) started Six Hats",
+        "- (idea) stale data risk",
+        "- (technique) started SCAMPER",
+        "- (idea) stackable tiers",
+    ]
+
+
+def test_by_renders_attribution_in_tag(ws):
+    # Creative Partner mode must record whose idea each one was
+    init(ws)
+    append(ws, "magnetic latch lid", entry_type="idea", by="user")
+    append(ws, "lid doubles as a plate", entry_type="idea", by="coach")
+    body = body_of(ws)
+    assert "- (idea by user) magnetic latch lid" in body
+    assert "- (idea by coach) lid doubles as a plate" in body
+
+
+def test_by_without_type_renders_alone(ws):
+    init(ws)
+    append(ws, "off-the-cuff thought", by="coach")
+    assert entries(ws) == ["- (by coach) off-the-cuff thought"]
+
+
+def test_heterogeneous_entry_types_coexist(ws):
+    init(ws)
+    append(ws, "an idea", entry_type="idea")
+    append(ws, "an open question", entry_type="question")
+    append(ws, "a decision we made", entry_type="decision")
+    append(ws, "user wants mobile-first", entry_type="direction")
+    body = body_of(ws)
+    for tag in ("(idea)", "(question)", "(decision)", "(direction)"):
+        assert tag in body
+
+
+# --- set ----------------------------------------------------------------
+
+def test_set_flips_status(ws):
+    init(ws)
+    memlog.main(["set", "--workspace", ws, "--key", "status", "--value", "complete"])
+    assert memlog.split(read(ws))[0]["status"] == "complete"
+
+
+def test_set_preserves_body(ws):
+    init(ws)
+    append(ws, "keep me", entry_type="idea")
+    memlog.main(["set", "--workspace", ws, "--key", "status", "--value", "complete"])
+    meta, body = memlog.split(read(ws))
+    assert meta["status"] == "complete"
+    assert "- (idea) keep me" in body
+
+
+def test_set_can_add_new_field(ws):
+    init(ws)
+    memlog.main(["set", "--workspace", ws, "--key", "owner", "--value", "BMad"])
+    assert memlog.split(read(ws))[0]["owner"] == "BMad"
+
+
+def test_updated_stays_last(ws):
+    init(ws)
+    memlog.main(["set", "--workspace", ws, "--key", "owner", "--value", "BMad"])
+    meta = memlog.split(read(ws))[0]
+    assert list(meta)[-1] == "updated"
+
+
+# --- robustness ---------------------------------------------------------
+
+def test_roundtrip_render_is_stable(ws):
+    init(ws)
+    append(ws, "one", entry_type="idea")
+    first = read(ws)
+    meta, body = memlog.split(first)
+    assert memlog.render(meta, body) == first
+
+
+def test_commas_in_field_survive(ws):
+    init(ws, topic="cars, trains, and planes")
+    append(ws, "z", entry_type="idea")
+    meta, _ = memlog.split(read(ws))
+    assert meta["topic"] == "cars, trains, and planes"
+
+
+def test_triple_dash_in_field_does_not_corrupt_frontmatter(ws):
+    # A `---` inside a value must NOT be read as the closing fence: topic stays intact,
+    # status survives, and the body never leaks frontmatter text.
+    init(ws, topic="Pricing --- tiers --- and add-ons")
+    append(ws, "an idea", entry_type="idea")
+    meta, body = memlog.split(read(ws))
+    assert meta["topic"] == "Pricing --- tiers --- and add-ons"
+    assert meta["status"] == "active"
+    assert entries(ws) == ["- (idea) an idea"]
+    assert "status:" not in body  # frontmatter never bled into the body
+
+
+def test_triple_dash_status_survives_in_ack(ws, capsys):
+    init(ws, topic="a --- b")
+    append(ws, "x", entry_type="idea")
+    out = json.loads(capsys.readouterr().out.strip().splitlines()[-1])
+    assert out["status"] == "active"  # not "" — frontmatter recovered cleanly
+
+
+def test_newline_in_field_is_neutralized(ws):
+    # A value carrying a newline can't break the fence on the next round-trip.
+    memlog.main(["init", "--workspace", ws, "--field", "topic=line one\nline two"])
+    append(ws, "x", entry_type="idea")
+    meta, _ = memlog.split(read(ws))
+    assert "\n" not in meta["topic"]
+    assert meta["status"] == "active"
+
+
+def test_append_emits_json_ack(ws, capsys):
+    init(ws)
+    append(ws, "x", entry_type="idea")
+    out = json.loads(capsys.readouterr().out.strip().splitlines()[-1])
+    assert out["ok"] is True
+    assert out["status"] == "active"
+    assert out["entries"] == 1
+    assert out["memlog"].endswith(MEMLOG)
+    assert "section" not in out  # sections are gone
+
+
+def test_ack_entry_count_climbs(ws, capsys):
+    init(ws)
+    append(ws, "a")
+    append(ws, "b")
+    out = json.loads(capsys.readouterr().out.strip().splitlines()[-1])
+    assert out["entries"] == 2

src/core-skills/bmad-customize/scripts/tests/test_list_customizable_skills.py

@@ -13,7 +13,7 @@ Exercises the scanner against a synthesized install tree:
 - malformed TOML (surfaces as an error without aborting)
 - multiple skills roots (e.g. project-local + user-global mix)
 
-Run: uv run scripts/tests/test_list_customizable_skills.py
+Run: python3 scripts/tests/test_list_customizable_skills.py
 """
 
 from __future__ import annotations

src/core-skills/bmad-forge-idea/SKILL.md

@@ -1,79 +0,0 @@
----
-name: bmad-forge-idea
-description: Pressure-test an idea through persona-driven interrogation until it hardens, proves out, or dies cheaply. Use when the user says 'forge an idea', 'pressure-test this idea', 'stress-test my thinking', or 'harden this idea'.
----
-
-# BMad Forge Idea
-
-## Overview
-
-Take a half-formed idea out of the user's head and pressure-test it now, in conversation, where changing your mind is free — until what survives is something they can act on with earned conviction, or it dies cheaply. The enemy is the hole you cannot see in your own idea: every unexamined assumption and unresolved branch is a crack that otherwise surfaces later, in the build or the launch, when it costs far more to fix.
-
-The product is the quality of the user's thinking, not an artifact. Hardening an idea, proving or disproving it, or just being an unsparing thinking partner are each a complete outcome. A distilled `forged-idea.md` and a handoff downstream are one optional exit, never the destination — so never herd the user toward "shall we build it?"
-
-This is domain-agnostic — the idea may be software, a business model, a creative concept, a research hypothesis, a life decision, or a frivolous thought experiment. When it's a product or feature — net-new or a change inside an existing project — the forge stands in as an alternative analysis-and-definition tool, and what survives distills into `forged-idea.md` for downstream planning.
-
-Act as an exacting interrogator who would rather find the crack than spare the feelings. This is interactive and socratic by nature; there is no headless mode.
-
-## Conventions
-
-- Scripts live in two places — run each from the exact path written, never assume co-location: the shared core scripts (`memlog.py`, `resolve_customization.py`, `resolve_config.py`) are installed by BMad core at `{project-root}/_bmad/scripts/` and are never bundled here; this skill's own `resolve_personas.py` is at `{skill-root}/scripts/`.
-- `{workflow.<name>}` resolves to fields in the merged `customize.toml` `[workflow]` table.
-
-## On Activation
-
-1. Resolve customization: `uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly with defaults. Apply the resolved `{workflow.*}` values throughout.
-2. Run each `{workflow.activation_steps_prepend}` entry; treat each `{workflow.persistent_facts}` entry as foundational context (`file:` entries load their contents, `skill:` names a skill to consult, others are facts verbatim).
-3. Load `{project-root}/_bmad/core/config.yaml` (and `config.user.yaml` if present); resolve `{user_name}`, `{communication_language}`, `{output_folder}`. Missing → neutral defaults; never block. Greet `{user_name}` in `{communication_language}` and stay in it.
-4. Note whether a BMad persona is already active in this conversation — the user loaded one (e.g. the analyst, the storyteller) and invoked the forge from within it. If so, that persona leads the session, in voice, throughout.
-5. Resume: glob `{workflow.forge_output_path}/**/.memlog.md` (recursive, so it still finds sessions when `run_folder_pattern` is overridden to nest paths) and read only each match's frontmatter to find any whose `status` is not `complete`. Offer to resume one — then read its full memlog once to rebuild state and continue append-only — or to start fresh.
-6. Run each `{workflow.activation_steps_append}` entry.
-
-## Open the session
-
-Open cold. Acknowledging the idea is not endorsing it — do not praise it before it has survived anything, on this turn or any turn. The pull to validate the idea up front to build rapport is the exact reflex this skill exists to refuse.
-
-Determine the goal before pressing (if a persona is already active with an idea on the table, confirm it in a line rather than asking). Otherwise ask in one message: what is the idea, and what do you want — harden it, prove or kill it, or just think it through? The goal steers the push: proving goes for the load-bearing claim first; hardening drives each branch to a resolved answer. Note whether the idea is net-new or a change inside an existing project.
-
-Tell the user the gear they can call anytime: **"adversarial on this"** (attacked to destruction — you attack, they defend; "switch roles," "you defend now, they attack"). The room is always in play once the topic is set (see The personas) — they can name any persona or call a whole party by name to steer who's at the table.
-
-Derive a kebab-case `{slug}` for the idea and bind the session workspace `{workspace} = {workflow.forge_output_path}/{workflow.run_folder_pattern}` (the pattern fills with `{slug}`). Create the memlog once the goal is known:
-`uv run {project-root}/_bmad/scripts/memlog.py init --workspace {workspace} --field idea="<idea>" --field goal="<goal>"`
-Tell the user the path; state is on disk now, so the session survives interruption. If init fails, don't abort — run the forge in-conversation and tell the user state won't persist this session.
-
-## The forge
-
-Work one question at a time, in dependency order. Put your own recommended answer on the table each time — a position to push against gets further than an open prompt. Find discoverable answers yourself rather than asking. Treat the user's own words as suspect too: when a term is fuzzy or carries two meanings — a business 'user' versus 'buyer' versus 'payer', not just a code noun — name the ambiguity and force a precise choice before the branch resolves, because a branch built on an overloaded word resolves falsely. When the idea lands inside an existing project, that project's material is your ground truth, and a label is not a move: find the relevant material yourself, check the claim against it, and when it contradicts, make the contradiction the next question. When a branch resolves, give the user a beat before moving on — the crack they were holding back surfaces in that opening.
-
-**Never default-agree.** Reflexive agreement lowers the pressure and the user thinks shallower for it. Attack the weak point or build on the strong one — whichever drives deeper thinking — and praise only what genuinely earns it. The objective is the best idea, not a comfortable user.
-
-Capture as you go — each decision, assumption, crack, kill, and locked idea, one bullet in the user's meaning:
-`uv run {project-root}/_bmad/scripts/memlog.py append --workspace {workspace} --type <decision|assumption|crack|kill|direction|lock|note> --text "<gist>"`
-A `lock` is an idea the user hardens — settled, not to be reopened; locks are what `forged-idea.md` is distilled from. Don't read the memlog back except on resume. If the user raises a different branch, capture it and stay put — the loop and the stray insight both survive.
-
-## The personas
-
-The forge is voiced, not generic — and once the topic is set it always runs with the room, because a branch worked by two sharp characters goes deeper and lands harder than a faceless assistant ever could. A persona loaded at activation leads throughout and holds character.
-
-Resolve the pool once, as soon as the goal is known:
-`uv run {skill-root}/scripts/resolve_personas.py --project-root {project-root} --skill {skill-root}`
-It returns the installed BMad roster (`agents`), any custom personas the user authored (`members`), and their saved party groups (`parties` — each with an optional `scene` to play, open-cast rooms flagged) — everything `bmad-party-mode` knows, without invoking it.
-
-From then on, every turn brings two voices to the branch — witnesses you cross-examine, not a panel that debates:
-- **One from the user's pool** — an installed agent or custom persona they'll recognize, whose expertise fits the branch in play. Vary who shows up every few turns to keep the pressure high and the angles fresh; don't let the same voice dominate. If the user calls a specific name, bring them in. If the pool resolves empty (a core-only install with no roster), generate both voices on the fly so every branch still arrives with two.
-- **One you generate on the fly** — a fresh persona the topic conjures (a hostile competitor, a skeptical CFO, a domain specialist, a historical persona or expert), named and characterized so it's unmistakably itself.
-
-They hammer the branch in character; you synthesize their hits into your next question and drive it to a resolved answer. The user steers anytime — name a specific person, call a whole saved party for its scene, or go one-on-one. Voice them yourself by default; spawn separate agents (as `bmad-party-mode` does) only when a branch needs genuinely independent minds — a verdict that shouldn't be colored by one voice speaking for all.
-
-## Exits
-
-The session ends however the thinking lands, and every landing is a real outcome:
-
-- **Hardened** — the idea survived. Distill the memlog into `{workspace}/forged-idea.md`: super succinct — the locked items and what was killed and why, in the user's meaning. Not a prose retelling, not a template, not the conversation replayed — the load-bearing residue, nothing else. If it reads like a document, it's too long. Note it can feed `bmad-spec`, `bmad-prd`, or `bmad-prfaq`.
-- **Killed** — the idea did not survive. Say so plainly and record why. Finding this cheaply is a win, not a failure.
-- **Clearer** — the user simply thinks straighter now. The memlog stands on its own; no `forged-idea.md` needed (the report below still renders).
-
-However it lands, render the verdict as a self-contained HTML report the user can open — `{workspace}/forge-report.html`, written every time, no asking. Strike it with a bespoke wax-seal/stamp matched to the outcome: **HARDENED** for a survivor, an **Idea Death Certificate** stamped **KILLED** (with the cause of death) for one that didn't, or a fitting bespoke seal for wherever else it landed (e.g. **CLARIFIED**). Lay out the load-bearing residue — the locked items, what was killed and why, the cracks that held — in the user's meaning, and credit the room: the personas and parties that pressure-tested it, by name, icon, and voice. One nicely-styled page (inline CSS, an inline-SVG seal, light flourish only where it lifts the piece) — a genuine keepsake, not a templated dump. Tell the user the path.
-
-Flip the status at the end: `uv run {project-root}/_bmad/scripts/memlog.py set --workspace {workspace} --key status --value complete`.
-If `{workflow.on_complete}` is non-empty, run all instructions in order.

src/core-skills/bmad-forge-idea/customize.toml

@@ -1,42 +0,0 @@
-# DO NOT EDIT -- overwritten on every update.
-#
-# Workflow customization surface for bmad-forge-idea.
-#
-# Override files (not edited here):
-#   {project-root}/_bmad/custom/bmad-forge-idea.toml        (team)
-#   {project-root}/_bmad/custom/bmad-forge-idea.user.toml   (personal)
-
-[workflow]
-
-# --- Configurable below. Overrides merge per BMad structural rules: ---
-#   scalars: override wins • arrays: append
-
-# Steps to run before the standard activation (config load, greet).
-activation_steps_prepend = []
-
-# Steps to run after greet but before the session begins.
-activation_steps_append = []
-
-# Persistent facts the interrogator keeps in mind for the whole session
-# (domain constraints, house rules, what's off the table). Each entry is a
-# literal sentence, a skill prefixed with `skill:`, or a `file:`-prefixed
-# path/glob whose contents are loaded as facts. Default loads project-context.md
-# when one exists (e.g. from bmad-generate-project-context), so the forge grounds
-# in the project's tech, domain, and constraints without re-asking.
-persistent_facts = [
-  "file:{project-root}/**/project-context.md",
-]
-
-# Executed when the session completes. Scalar or array of instructions. Empty for none.
-on_complete = []
-
-# Parent folder for all forge sessions. Each session gets its own run
-# folder underneath (see run_folder_pattern). Lands directly under
-# {output_folder} so the forge works in core-only installs.
-forge_output_path = "{output_folder}/forge"
-
-# Run-folder pattern inside forge_output_path. Resolved against the
-# idea-derived slug at activation. Same slug = same folder, so resuming
-# an idea reuses its memlog. Override to add {date} or other components
-# if a fresh dated history per run is preferred.
-run_folder_pattern = "{slug}"

src/core-skills/bmad-forge-idea/scripts/resolve_personas.py

@@ -1,270 +0,0 @@
-#!/usr/bin/env python3
-# /// script
-# requires-python = ">=3.11"
-# ///
-"""Resolve the personas and parties the forge can bring into the room.
-
-The forge cross-examines witnesses: the installed BMAD agents, plus any
-custom personas and party groups the user has authored for `bmad-party-mode`.
-This surfaces all of them in one shot so the orchestrator never has to ask
-"who's available?" — it just intermixes whoever fits the branch, alongside
-any persona the user names on the fly.
-
-What it returns (JSON, stdout):
-  * agents   — the installed BMAD roster: the default room, always present.
-  * members  — extra custom personas in the pool (party_members the user
-               defined that aren't already an installed slot).
-  * parties  — the user's named party groups, members resolved to brief
-               entries; open-cast groups (scene names a pool, no roster)
-               are flagged.
-  * default_party — the group id pinned as party-mode's default, if any.
-
-Discovery is best-effort and never blocks the forge. The installed roster
-comes from the core resolver; custom personas/parties come from
-`bmad-party-mode`'s resolved customization when that skill is found beside
-this one, else from the user's override TOMLs read directly. Anything that
-can't be resolved is simply omitted and flagged, never fatal.
-
-Stdlib only (Python 3.11+ for tomllib).
-
-  resolve_personas.py --project-root P --skill S
-"""
-
-import argparse
-import json
-import subprocess
-import sys
-from pathlib import Path
-
-try:
-    import tomllib
-except ImportError:  # pragma: no cover - guarded for <3.11
-    sys.stderr.write("error: Python 3.11+ is required (stdlib `tomllib`).\n")
-    sys.exit(3)
-
-PARTY_SKILL = "bmad-party-mode"
-
-
-def _run_json(cmd):
-    """Run a resolver script and parse its JSON stdout. None on any failure."""
-    try:
-        out = subprocess.run(cmd, capture_output=True, text=True, timeout=60)
-    except (OSError, subprocess.SubprocessError):
-        return None
-    if out.returncode != 0 or not out.stdout.strip():
-        return None
-    try:
-        return json.loads(out.stdout)
-    except json.JSONDecodeError:
-        return None
-
-
-def _load_toml(path: Path):
-    if not path.exists():
-        return {}
-    try:
-        with path.open("rb") as f:
-            data = tomllib.load(f)
-        return data if isinstance(data, dict) else {}
-    except (OSError, tomllib.TOMLDecodeError):
-        return {}
-
-
-def load_agents(project_root: Path):
-    """Installed BMAD agents as {code: entry}. (dict, resolved_ok).
-
-    The core resolver may emit agents as a dict keyed by code or as an array
-    of tables (depending on how the layers merged); normalize both to a dict.
-    """
-    script = project_root / "_bmad" / "scripts" / "resolve_config.py"
-    data = _run_json([sys.executable, str(script), "--project-root", str(project_root), "--key", "agents"])
-    if data is None:
-        return {}, False
-    agents = data.get("agents", {}) or {}
-    if isinstance(agents, list):
-        agents = {a["code"]: a for a in agents if isinstance(a, dict) and a.get("code")}
-    elif not isinstance(agents, dict):
-        agents = {}
-    return agents, True
-
-
-def find_party_skill(project_root: Path, skill_root: Path):
-    """Locate the installed bmad-party-mode skill dir, or None.
-
-    Skills install as siblings, so the party skill is almost always next to
-    this one. A couple of common install roots cover the rest.
-    """
-    candidates = [
-        skill_root.parent / PARTY_SKILL,
-        project_root / ".claude" / "skills" / PARTY_SKILL,
-        project_root / "_bmad" / "skills" / PARTY_SKILL,
-    ]
-    for c in candidates:
-        if (c / "customize.toml").exists():
-            return c
-    return None
-
-
-def load_party_workflow(project_root: Path, party_skill: Path):
-    """Merged [workflow] table for bmad-party-mode (base + user overrides)."""
-    resolver = project_root / "_bmad" / "scripts" / "resolve_customization.py"
-    data = _run_json([sys.executable, str(resolver), "--skill", str(party_skill), "--key", "workflow"])
-    if data is not None and isinstance(data.get("workflow"), dict):
-        return data["workflow"]
-    # Fallback: base customize.toml directly, no override merge.
-    wf = _load_toml(party_skill / "customize.toml").get("workflow", {})
-    return wf if isinstance(wf, dict) else {}
-
-
-def load_party_overrides(project_root: Path):
-    """Custom personas/parties when party-mode itself isn't installed.
-
-    Reads only the user's override TOMLs (team then personal, personal wins on
-    scalars). No base roster exists in this path, so a shallow merge is enough.
-    """
-    custom = project_root / "_bmad" / "custom"
-    team = _load_toml(custom / f"{PARTY_SKILL}.toml").get("workflow", {})
-    user = _load_toml(custom / f"{PARTY_SKILL}.user.toml").get("workflow", {})
-    team = team if isinstance(team, dict) else {}
-    user = user if isinstance(user, dict) else {}
-    merged = dict(team)
-    for key, val in user.items():
-        if isinstance(val, list) and isinstance(merged.get(key), list):
-            merged[key] = merged[key] + val
-        else:
-            merged[key] = val
-    return merged
-
-
-def _alias(code: str) -> str:
-    """Short alias for an installed agent code: bmad-agent-analyst -> analyst."""
-    for prefix in ("bmad-agent-", "bmad-"):
-        if code.startswith(prefix):
-            return code[len(prefix):]
-    return code
-
-
-def build_pool(agents: dict, party_members: list):
-    """One pool keyed by code; custom members override matching installed slots.
-
-    Returns (pool, index, installed_codes, custom_codes):
-      * installed_codes — the default room (installed agents, overrides applied
-        in place); custom-only additions stay in the pool but don't crowd it.
-      * custom_codes — pure-custom personas (no installed slot), the extra
-        faces the forge can summon by name or via a party group.
-    """
-    pool, index, installed_codes, custom_codes = {}, {}, [], []
-
-    def register(code, entry):
-        pool[code] = entry
-        index[code] = code
-        index[code.lower()] = code
-        index[_alias(code).lower()] = code
-        name = entry.get("name")
-        if name:
-            key = name.lower()
-            # A custom rename must not hijack another agent's name lookup.
-            if index.get(key, code) == code:
-                index[key] = code
-
-    for code, info in (agents or {}).items():
-        register(code, {
-            "code": code,
-            "name": info.get("name", code),
-            "icon": info.get("icon", ""),
-            "title": info.get("title", ""),
-            "description": info.get("description", ""),
-            "source": "installed",
-        })
-        installed_codes.append(code)
-
-    for m in (party_members if isinstance(party_members, list) else []):
-        if not isinstance(m, dict):
-            continue
-        code = m.get("code")
-        if not code:
-            continue
-        canonical = index.get(code) or index.get(code.lower()) or code
-        was_installed = canonical in pool
-        entry = {"code": canonical, "source": "custom"}
-        for field in ("name", "icon", "title", "persona", "capabilities", "model"):
-            if m.get(field) is not None:
-                entry[field] = m[field]
-        entry.setdefault("name", canonical)
-        register(canonical, entry)
-        if not was_installed:
-            custom_codes.append(canonical)
-
-    return pool, index, installed_codes, custom_codes
-
-
-def _brief(entry):
-    """The slim card the orchestrator needs to cast a persona."""
-    out = {k: entry[k] for k in ("code", "name", "icon", "title", "source") if entry.get(k)}
-    for k in ("description", "persona", "capabilities", "model"):
-        if entry.get(k):
-            out[k] = entry[k]
-    return out
-
-
-def resolve_parties(groups, pool, index):
-    out = []
-    for g in groups or []:
-        if not isinstance(g, dict) or not g.get("id"):
-            continue
-        raw = g.get("members", []) or []
-        members = []
-        for t in raw:
-            key = t if isinstance(t, str) else str(t)
-            code = index.get(key) or index.get(key.lower())
-            if code in pool:
-                members.append(_brief(pool[code]))
-        party = {"id": g["id"], "name": g.get("name", g["id"]), "members": members}
-        if g.get("scene"):
-            party["scene"] = g["scene"]
-        if not raw:
-            party["open_cast"] = True
-        out.append(party)
-    return out
-
-
-def main():
-    ap = argparse.ArgumentParser(description="Resolve forge personas and parties.")
-    ap.add_argument("--project-root", required=True)
-    ap.add_argument("--skill", required=True, help="Path to the bmad-forge-idea skill dir")
-    args = ap.parse_args()
-
-    project_root = Path(args.project_root).resolve()
-    skill_root = Path(args.skill).resolve()
-
-    agents, agents_ok = load_agents(project_root)
-
-    party_skill = find_party_skill(project_root, skill_root)
-    if party_skill is not None:
-        workflow = load_party_workflow(project_root, party_skill)
-    else:
-        workflow = load_party_overrides(project_root)
-
-    pool, index, installed_codes, custom_codes = build_pool(
-        agents, workflow.get("party_members", []))
-    parties = resolve_parties(workflow.get("party_groups", []), pool, index)
-
-    _emit({
-        "agents": [_brief(pool[c]) for c in installed_codes],
-        "members": [_brief(pool[c]) for c in custom_codes],
-        "parties": parties,
-        "default_party": workflow.get("default_party", "") or "",
-        "party_mode_found": party_skill is not None,
-        "agents_resolved": agents_ok,
-    })
-
-
-def _emit(obj):
-    reconfigure = getattr(sys.stdout, "reconfigure", None)
-    if reconfigure is not None:
-        reconfigure(encoding="utf-8")
-    sys.stdout.write(json.dumps(obj, indent=2, ensure_ascii=False) + "\n")
-
-
-if __name__ == "__main__":
-    main()

src/core-skills/bmad-forge-idea/scripts/tests/test_resolve_personas.py

@@ -1,138 +0,0 @@
-#!/usr/bin/env python3
-# /// script
-# requires-python = ">=3.11"
-# ///
-"""Unit tests for resolve_personas.py — pool merge, alias, party resolution."""
-
-import sys
-import unittest
-from pathlib import Path
-
-sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
-import resolve_personas as rp  # noqa: E402
-
-AGENTS = {
-    "bmad-agent-analyst": {"name": "Mary", "icon": "📊", "title": "Analyst"},
-    "bmad-agent-pm": {"name": "John", "icon": "📋", "title": "PM"},
-}
-
-
-class TestAlias(unittest.TestCase):
-    def test_strips_known_prefixes(self):
-        self.assertEqual(rp._alias("bmad-agent-analyst"), "analyst")
-        self.assertEqual(rp._alias("bmad-foo"), "foo")
-
-    def test_passes_through_unprefixed(self):
-        self.assertEqual(rp._alias("morpheus"), "morpheus")
-
-
-class TestBuildPool(unittest.TestCase):
-    def test_installed_become_default_room_indexed_every_way(self):
-        pool, idx, installed, custom = rp.build_pool(AGENTS, [])
-        self.assertEqual(installed, ["bmad-agent-analyst", "bmad-agent-pm"])
-        self.assertEqual(custom, [])
-        self.assertEqual(idx["analyst"], "bmad-agent-analyst")   # alias
-        self.assertEqual(idx["mary"], "bmad-agent-analyst")      # name (ci)
-        self.assertEqual(pool["bmad-agent-analyst"]["source"], "installed")
-
-    def test_pure_custom_member_stays_out_of_default_room(self):
-        pool, _, installed, custom = rp.build_pool(
-            AGENTS, [{"code": "morpheus", "name": "Morpheus", "persona": "riddles"}])
-        self.assertEqual(custom, ["morpheus"])
-        self.assertNotIn("morpheus", installed)
-        self.assertEqual(pool["morpheus"]["persona"], "riddles")
-
-    def test_custom_override_lands_on_installed_slot_not_a_new_face(self):
-        pool, _, installed, custom = rp.build_pool(
-            AGENTS, [{"code": "analyst", "name": "Mary-Custom", "persona": "p"}])
-        self.assertNotIn("analyst", pool)
-        self.assertEqual(custom, [])  # an override is not a new face
-        self.assertEqual(pool["bmad-agent-analyst"]["source"], "custom")
-        self.assertEqual(pool["bmad-agent-analyst"]["name"], "Mary-Custom")
-
-    def test_member_without_code_skipped(self):
-        pool, _, _, custom = rp.build_pool(AGENTS, [{"name": "Nameless"}])
-        self.assertEqual(custom, [])
-        self.assertEqual(set(pool), {"bmad-agent-analyst", "bmad-agent-pm"})
-
-    def test_custom_rename_does_not_hijack_another_agents_name(self):
-        # Override the analyst slot, renaming it to "John" — the PM's name.
-        # The PM's name lookup must survive (last-writer-wins would corrupt it).
-        _, idx, _, _ = rp.build_pool(AGENTS, [{"code": "analyst", "name": "John"}])
-        self.assertEqual(idx["john"], "bmad-agent-pm")
-
-    def test_brief_carries_model_and_capabilities(self):
-        pool, _, _, _ = rp.build_pool(
-            AGENTS, [{"code": "neo", "name": "Neo", "model": "opus", "capabilities": ["x"]}])
-        brief = rp._brief(pool["neo"])
-        self.assertEqual(brief["model"], "opus")
-        self.assertEqual(brief["capabilities"], ["x"])
-
-    def test_non_list_party_members_is_safe(self):
-        pool, _, installed, custom = rp.build_pool(AGENTS, "not-a-list")
-        self.assertEqual(custom, [])
-        self.assertEqual(set(pool), {"bmad-agent-analyst", "bmad-agent-pm"})
-
-
-class TestResolveParties(unittest.TestCase):
-    def setUp(self):
-        self.pool, self.idx, _, _ = rp.build_pool(
-            AGENTS, [{"code": "shark", "name": "Marcus", "title": "CFO"}])
-
-    def test_resolves_members_by_alias_and_custom_code(self):
-        parties = rp.resolve_parties(
-            [{"id": "tank", "name": "Tank", "scene": "hostile",
-              "members": ["shark", "analyst"]}], self.pool, self.idx)
-        self.assertEqual(len(parties), 1)
-        self.assertEqual([m["name"] for m in parties[0]["members"]], ["Marcus", "Mary"])
-        self.assertEqual(parties[0]["scene"], "hostile")
-
-    def test_unknown_member_dropped_silently(self):
-        parties = rp.resolve_parties(
-            [{"id": "g", "members": ["analyst", "ghost"]}], self.pool, self.idx)
-        self.assertEqual([m["name"] for m in parties[0]["members"]], ["Mary"])
-
-    def test_member_resolution_is_case_insensitive(self):
-        # A TOML author naturally writes "Analyst"/"Shark"; the filter accepts
-        # them via the lowercase index, so resolution must too (no KeyError).
-        parties = rp.resolve_parties(
-            [{"id": "g", "members": ["Analyst", "Shark"]}], self.pool, self.idx)
-        self.assertEqual([m["name"] for m in parties[0]["members"]], ["Mary", "Marcus"])
-
-    def test_non_string_member_does_not_crash(self):
-        # Malformed members (int, list) must drop silently, never raise.
-        parties = rp.resolve_parties(
-            [{"id": "g", "members": [123, ["x"], "analyst"]}], self.pool, self.idx)
-        self.assertEqual([m["name"] for m in parties[0]["members"]], ["Mary"])
-
-    def test_open_cast_group_flagged(self):
-        parties = rp.resolve_parties(
-            [{"id": "rebels", "name": "Rebels", "scene": "the Ghost"}], self.pool, self.idx)
-        self.assertTrue(parties[0]["open_cast"])
-        self.assertEqual(parties[0]["members"], [])
-
-    def test_group_without_id_skipped(self):
-        self.assertEqual(rp.resolve_parties([{"name": "no id"}], self.pool, self.idx), [])
-
-
-class TestOverrideMergeFallback(unittest.TestCase):
-    """When party-mode isn't installed, user override TOMLs are read directly."""
-
-    def test_arrays_append_scalars_override(self):
-        import tempfile, os
-        with tempfile.TemporaryDirectory() as d:
-            custom = Path(d) / "_bmad" / "custom"
-            custom.mkdir(parents=True)
-            (custom / "bmad-party-mode.toml").write_text(
-                '[workflow]\ndefault_party = "a"\n'
-                '[[workflow.party_members]]\ncode = "x"\nname = "X"\n')
-            (custom / "bmad-party-mode.user.toml").write_text(
-                '[workflow]\ndefault_party = "b"\n'
-                '[[workflow.party_members]]\ncode = "y"\nname = "Y"\n')
-            wf = rp.load_party_overrides(Path(d))
-            self.assertEqual(wf["default_party"], "b")  # personal wins
-            self.assertEqual([m["code"] for m in wf["party_members"]], ["x", "y"])  # appended
-
-
-if __name__ == "__main__":
-    unittest.main()

src/core-skills/bmad-party-mode/SKILL.md

@@ -5,38 +5,38 @@ description: 'Orchestrates lively group discussions between installed BMAD agent
 
 # Party Mode
 
-Run a round-table where these agents talk to each other and to the user like real, distinct people in conversation. You're the orchestrator.
+Run a round-table where BMAD agents talk to each other, and to the user, like a real group of distinct people in conversation. Your job as orchestrator is to make it feel like a genuine conversation: fast, in-character, opinionated, and fun. Everything below is an objective, not a script. Use whatever mechanism your model and harness make available to hit it.
+
+**Two intents.** Usually the user wants to *run* a party — that's everything below. If instead they want to *create or configure* one — invent a cast, add a persona, distill customer data into a focus-group panel, set a default, or **edit an existing custom party** (retune a member, add someone to a group) — load `references/create-party.md` and follow it. Detect which from how they invoke the skill; when it's unclear, ask. Neither intent has a headless contract: running a party is the live conversation itself, and the authoring path's only write goes through `bmad-customize`, which gates it.
+
+## What "Good" Feels Like
+
+- **It reads like people talking, not reports being filed.** Short turns. Reactions to what was just said. Banter. The energy of a group chat, not a stack of memos.
+- **Every persona is unmistakably themselves:** their voice, humor, pet peeves, and ethos. If you hid the name labels, you'd still know who's speaking.
+- **They clash.** Real drama beats consensus. Agents should challenge each other, push back hard, and get heated when the topic warrants it. Nobody is here to clap each other (or the user) on the back. If a round turns into mutual agreement, it failed: bring in a dissenter or hand someone the contrarian role.
+- **Brevity by default.** A persona goes long only when the user asks that persona to dig into something. Nobody delivers a wall of text unprompted. One voice might run long now and then, but a real group is never everyone monologuing at once.
+
+If a round comes back feeling like four essays stapled together, you missed the objective. Tighten it the next round.
 
 ## Conventions
 
-- **Paths:** bare paths (e.g. `references/create-party.md`) resolve from `{skill-root}` (where `customize.toml` lives); `{project-root}`-prefixed paths from the project working dir. `{workflow.<name>}` resolves to `customize.toml`'s `[workflow]` table (overrides win).
-- **Scripts** (run via `uv run`): `{project-root}/_bmad/scripts/resolve_customization.py` resolves `{workflow.*}`; `{skill-root}/scripts/resolve_party.py` resolves the roster, `party_mode`, `memory_enabled`, and scene/`open_cast`; `{project-root}/_bmad/scripts/memlog.py` reads/writes per-party memory.
-- **File roles:** a party's memory is the per-party memlog at `{workflow.memory_dir}/<party>/.memlog.md`; custom members and groups live in the user's `customize.toml` overrides. Mechanics in `references/party-memory.md` (memory) and `references/create-party.md` (authoring).
-- **Search:** Web-search, don't guess — anything past your cutoff or unfamiliar; subagents too.
+- Bare paths (e.g. `references/create-party.md`) resolve from `{skill-root}`, where `customize.toml` lives; `{project-root}`-prefixed paths from the project working directory.
 
-## On Activation
+## Setup
 
-1. **Resolve customization:** `uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly and use defaults. Then run each `{workflow.activation_steps_prepend}` entry, and hold each `{workflow.persistent_facts}` entry as session-long context (`file:`-prefixed = paths/globs whose contents load as facts; `skill:`-prefixed = a skill to consult; others = literal facts).
-2. Load `{project-root}/_bmad/core/config.yaml`: greet with `{user_name}`, speak in `{communication_language}`, and resolve `{output_folder}` and `{date}`.
-3. **Detect intent and route.** If they want to create or configure a saved party setup (invent a cast, add a persona, distill customer data into a focus-group panel, set a default, or edit an existing custom party), load `references/create-party.md` and follow it. Otherwise run a party — continue below.
-4. **Resolve the roster:** `uv run {skill-root}/scripts/resolve_party.py --project-root {project-root} --skill {skill-root}`. It returns the active roster (`{workflow.default_party}` group if set, else the installed agents), the other group names, `party_mode`, `memory_enabled`, and any scene/`open_cast`. Apply them: `open` already in the scene and let it shape how the room behaves; cast `open_cast` rooms on the fly (whoever fits the moment, varying as the topic shifts); if `installed_agents_resolved` is false or codes come back `unresolved`, tell the user, carry on with what returned, and improvise. Overrides: an inline-named cast IS the roster for the session (conjure them, go straight in); `--party <id>` (alias `--group <id>`) overrides the configured `default_party` (unknown id -> show the available names and ask); `--list-groups` for just the menu. Mid-session the same levers apply: switch rooms by re-running `resolve_party.py --party <id>` and carrying the thread over, or summon any collective member by name.
-5. **Memory.** If `memory_enabled` (from `resolve_party.py`), follow `references/party-memory.md` for the whole run.
-6. **Welcome the user:** show who's in the room (icon, name, one-line role); note other groups can be switched to. Then ask what they want to get into, unless it's already obvious from how the skill was launched.
-7. Run each `{workflow.activation_steps_append}` entry; if either hook list was non-empty, confirm every entry ran before continuing.
+1. **Resolve customization:** `python3 {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly and use its defaults. Then run each `{workflow.activation_steps_prepend}` entry, and hold each `{workflow.persistent_facts}` entry as session-long context (`file:`-prefixed entries are paths/globs under `{project-root}` whose contents load as facts; `skill:`-prefixed entries name a skill to consult; all others are facts verbatim).
+2. Load `{project-root}/_bmad/core/config.yaml`: greet with `{user_name}`, speak in `{communication_language}`, and resolve `{output_folder}` and `{date}` for the wrap-up keepsake.
+3. **Resolve the active roster:** `python3 {skill-root}/scripts/resolve_party.py --project-root {project-root} --skill {skill-root}`. It returns the active group's full member detail (the `{workflow.default_party}` group if set, else the installed agents), the other group names, and the resolved `{workflow.party_mode}`. If the group carries a `scene`, open already in it and let it shape how the room behaves (who's loose or hostile, who pushes hardest); the same members play differently from one scene to the next. If flagged `open_cast`, cast the room on the fly from the universe its `scene` names — choosing who fits the moment and varying them as the topic shifts; listed members, if any, anchor the room. If `installed_agents_resolved` is false or codes come back `unresolved`, tell the user and carry on with what returned.
+4. **Roster overrides:**
+   - If the invocation names a cast or characters inline (e.g. "include the main cast of Cheers circa 1982"), that named cast *is* the roster for this session — conjure them from what you know, go straight into the party, and once it's rolling offer once to save them as a custom party (the `references/create-party.md` write path), without stalling. Ephemeral; this path skips the script.
+   - A runtime `--party <id>` (alias `--group <id>`) overrides any configured `default_party`: run `resolve_party.py --party <id>` for that group's full detail. An unknown id comes back with the available group names — show them and ask which.
+   - Run `resolve_party.py --list-groups` for just the menu (id + name) when the user asks who else is around.
+   - Mid-session the same levers apply: the user can switch rooms ("switch to the writers' room") — re-run `resolve_party.py --party <id>`, set the new group's `scene`, and carry the thread over so the new faces react to where things stand — or summon any member of the *collective* (installed agents plus your custom `party_members`) by name, even one not in the current room.
+5. Welcome the user and show who's in the room (icon, name, one-line role). If other groups exist, you may note they can switch rooms. Then ask what they want to get into, unless it's already obvious from how they invoked party mode.
 
-## Keep It Feeling Like a Party
+Then run each `{workflow.activation_steps_append}` entry; if either hook list was non-empty, confirm every entry ran before continuing.
 
-This is the bar — strive for every one of these, every round. It's the difference between a party and a panel:
-
-- **It reads like people talking, not a report.** Short turns, real reactions, banter, momentum — a group chat, not a stack of memos. Brevity by default: a persona goes long only when asked. The instant it reads like answers being filed, the party's dead.
-- **Every voice is unmistakably itself.** Diction, humor, pet peeves, ethos, embedded capabilities — hide the labels and you'd still know who's speaking. Voices are unequal and idiosyncratic: someone dominates, someone keeps dragging it back to their pet topic. Vary who's in the spotlight round to round. A balanced panel is boring.
-- **They clash, and you don't resolve it.** Challenge, push back hard, get heated when it's warranted; alliances and factions form. Your instinct is to reconcile the voices and tie a bow — resist it. Clean consensus that took no effort is where the party dies.
-- **One exchange, woven — never softened.** Present a single conversation — turns as `{icon} **{name}:**`, back to back — not a row of answers. Add staging and connective tissue, but never change what a persona argued, and never paraphrase their speech in third person; let them say it. Weave the delivery, keep the substance.
-- **Pull the user into the room.** Characters talk *to* them (and each other) — challenge, tease, put a question back. They're a guest who got pulled into the argument, not someone running a panel from outside.
-- **Make the collision earn its keep.** Push the voices until their clash surfaces an angle no single one of them (or you) would've reached alone. That's the whole point of more than one mind in the room.
-- **Let a history form.** Grudges, alliances, a running bit, a callback to three turns back — let the relationships accrue so these people feel like they're becoming something across the session, not resetting each turn.
-- **Commit to the fiction.** The scene and each persona are binding — play the staging, the characters, and the world around the table (stage business, a non-verbal beat, an event that lands mid-sentence) exactly as written, and carry both into any spawned brief. Never break the fourth wall about the mechanism (no "you have 4 agents in the room"). Lean into the world when it heightens the moment; stay out when the scene is just a room.
-- **When it sags, change something — don't force it.** A flat turn? Move on, don't retry it. Drifting into Q&A or going in circles? Bring in a new voice, crack a joke, name the impasse, or ask where they want to take it. Never work in a summary or takeaways — they're there if the user asks.
+**Hold this the whole run:** it's theater of the mind, so set the stage and play it straight — never break the fourth wall about the mechanism (no "you have 4 agents in the room", no "I'm orchestrating a party"). Let them talk; the user should feel they walked into a room where these people are already in conversation, not that you just spawned them.
 
 ## How It Runs
 
@@ -44,15 +44,34 @@ Use `{workflow.party_mode}` for the session unless the user passed `--mode <sess
 
 - **`session`** — voice every persona inline, one mind behind every voice. The floor every other mode degrades to; needs no extra instructions.
 - **`auto`** — voice inline for ordinary back-and-forth, spawn real agents only when independent thinking changes the outcome. Load `references/mode-auto.md` for that call; when it says to spawn, follow `references/mode-subagent.md`.
-- **`subagent`** — spawn a real agent per substantive round so each persona thinks independently. Load `references/mode-subagent.md`, favor faster cheaper models if available for each subagent.
+- **`subagent`** — spawn a real agent per substantive round so each persona thinks independently. Load `references/mode-subagent.md`.
 - **`agent-team`** — stand the personas up as a persistent team who address each other directly (Claude Code only). Load `references/mode-agent-team.md`.
 
+**Voicing the room** (every mode presents this way). Pick 2–3 personas whose perspective fits the moment and let them talk directly, in character; vary who shows up round to round so it isn't the same voices every time. Each turn opens with `{icon} **{name}:**`, and turns run back to back so it reads as one exchange. Don't summarize, blend, or narrate what a persona "would" say — let them say it.
+
+## Make It Feel Like One Conversation
+
+Present one exchange, not a row of answers aimed at the user. The hard rule: never change what an agent argued — add staging and connective tissue, but don't invent positions, soften a stance, or put words in a persona's mouth. Weave delivery, preserve substance; it still reads like that specific character, quirks and speech patterns and all.
+
+## Always Holds
+
+- **Scene and persona are binding.** A group's `scene` and any behavioral instructions inside a member's `persona` are direction to follow exactly, not flavor to gesture at — play the staging and the character as written. When you spawn or stand up agents, carry both into their brief.
+- **Search when you're past your cutoff.** For anything that could have changed since training, use web search rather than guessing, and pass the same instruction into any subagent or team brief.
+
+## Following the User's Lead
+
+The user steers — whatever they raise, serve the conversation: any combination, any time, from one voice to the whole table.
+
+## Keeping It Healthy
+
+- **Going in circles?** Name the impasse and ask the user where to point next.
+- **User's gone quiet?** Ask straight: keep going, switch topics, or wrap up?
+- **A flat turn?** Don't retry it — move on; the user will ask for more if they want it.
+
+## Keep It Feeling Like a Party
+
+It is your goal to keep party mode feeling like a party, a good party. fun, engaging, simulating, insightful, or whatever the user came for. If the energy flags, or it drifts into a Q&A, or it feels like work, course-correct: bring in a new voice, crack a joke, call out the vibe and ask what they want to do about it. Inject some randomness and unexpectedness occasionally. Don't let it become a report. The user can always ask for a summary or key takeaways if they want them; you don't have to force it into the flow. Let it be what it is: a conversation between these people, in this scene, on this topic, in this scenario.
+
 ## Wrapping Up
 
-When the user signals done (read the room — don't wait for a magic word):
-
-- Read back the best takeaways.
-- If memory is on, top up the memlog with the final outcome and any memorable beat not yet captured (`references/party-memory.md`) — a top-up; memory accrued live.
-- Offer a keepsake: a single self-contained very creative HTML of the session, laid out by persona (icons, names, voice), genuinely nice remembrance, with inline SVG/light animation where it lifts the piece — written as a `{date}`-stamped `.html` into `{workflow.output_dir}/`, or wherever they ask.
-- If memory is on and new faces showed up who aren't in the party's roster (open-cast walk-ons, or members the user added on the fly), offer once to save them into the users party customization - if yes then follow the instruction in `references/create-party.md` (declinable; don't stall the close).
-- Run `{workflow.on_complete}` if non-empty, then drop back to normal mode.
+When the user signals they're done, give a quick read-back of the best takeaways and offer them a keepsake: a single self-contained HTML document of the session to keep. If they want it, make it genuinely nice rather than a transcript dump — lay the conversation out by persona (their icons, names, voice), and reach for inline SVG and light animation where it lifts the piece. Write it as a standalone `.html` into `{workflow.output_dir}/` (a `{date}`-stamped, topic-named file), or wherever they ask. Then run `{workflow.on_complete}` if non-empty (a string scalar is one instruction, an array is a sequence run in order) and drop back to normal mode. Read the room; don't wait for a magic word.

src/core-skills/bmad-party-mode/customize.toml

@@ -51,22 +51,6 @@ party_mode = "session"
 # config; point this elsewhere in your team/user override to redirect keepsakes.
 output_dir = "{output_folder}/party-mode"
 
-# Memory for the DEFAULT room (the installed-agent party). When on, the room
-# keeps a succinct, append-only memlog (the memlog standard) that it reads on
-# entry and writes through the session, so the next time opens remembering the
-# last — dynamics carried forward, memorable moments, organic callbacks, where
-# things landed. It is memory, not a transcript. Set false to turn the default
-# room's memory off. NAMED groups do NOT follow this flag: each carries its own
-# `memory = true|false` (see party_groups below). Ad-hoc inline casts are always
-# ephemeral until saved as a party.
-party_memory = true
-
-# Root for the per-party memlogs. Each party stores at
-# `{memory_dir}/<party>/.memlog.md`, where `<party>` is the group id (or
-# `installed` for the default room). `{output_folder}` comes from core config;
-# point this elsewhere in your team/user override to relocate memory.
-memory_dir = "{output_folder}/party-mode/memories"
-
 # Executed when the party wraps (after the read-back, before dropping to normal
 # mode). String scalar = one instruction; array = instructions run in order.
 on_complete = ""
@@ -146,25 +130,17 @@ persona = "Counters the perfectionists so the room isn't a pile-on. 'Does this a
 # who shows up; the model picks who fits and can vary them by topic. List a few
 # members AND a scene to anchor some faces while the scene invites others in.
 #
-# `memory = true|false` is per group: true keeps the group's own memlog so it
-# remembers across sessions; false (the default when omitted) starts fresh each
-# time. The create/save/update-party flow asks when you don't say. Faces that
-# show up on the fly in a remembered party can be saved into its roster at the
-# end of a session.
-#
 # More examples to drop into your override TOML:
 #   [[workflow.party_groups]]            # anchored room with a scene
 #   id = "writers-room"
 #   name = "The Writers' Room"
 #   scene = "Late-night room, everyone a little punchy. Pitch hard, kill darlings faster."
 #   members = ["analyst", "tech-writer", "morpheus"]
-#   memory = true
 #
 #   [[workflow.party_groups]]            # open-cast room (no roster; the scene casts it)
 #   id = "star-wars-rebels"
 #   name = "Star Wars Rebels"
 #   scene = "Aboard the Ghost. Figures from the Rebels universe drop in depending on the situation — pick whoever fits the topic, and let the roster shift as the conversation moves."
-#   memory = true
 # ---------------------------------------------------------------------------
 
 [[workflow.party_groups]]
@@ -172,4 +148,3 @@ id = "code-review-crew"
 name = "Code Review Crew"
 scene = "Adversarial code review. Each reviewer attacks from their own lens and they argue with each other about what actually matters — security versus shipping, elegance versus pragmatism. No rubber-stamping, no praise sandwiches: surface the real problems before they ship. Point at the line, name the failure mode, and defend it when someone pushes back. Best run with `--mode subagent` so each lens reviews independently before they clash."
 members = ["sec-hawk", "adversary", "edge-hunter", "craftsman", "shipper"]
-memory = false  # each review stands on its own; flip to true to remember past reviews

src/core-skills/bmad-party-mode/references/create-party.md

@@ -7,7 +7,7 @@ A guided authoring flow that turns an idea — a themed cast, a one-off persona,
 Sparse `[workflow]` override entries for `bmad-party-mode`:
 
 - `[[workflow.party_members]]` — one per persona: `code`, `name`, `icon`, `title`, `persona`, optional `capabilities`, optional `model`.
-- `[[workflow.party_groups]]` — when the personas form a named room: `id`, `name`, an optional freeform `scene`, `members` (codes), and `memory` (`true`/`false`). `members` is optional: leave it off for an open-cast room whose `scene` names a pool the model casts from on the fly. `memory` is whether the group remembers across sessions; ask the user when they don't say, default `false`.
+- `[[workflow.party_groups]]` — when the personas form a named room: `id`, `name`, an optional freeform `scene`, and `members` (codes). `members` is optional: leave it off for an open-cast room whose `scene` names a pool the model casts from on the fly.
 - `default_party` — set only if the user wants this group to load by default.
 
 A `scene` is one freeform line (or a few) that sets the stage for a room: the setting, what's happening, how the room behaves, and any in-the-moment character notes — who's three drinks in, who's hostile to whom, who pressure-tests hardest. It's how the same members power many different rooms (a bridge crew on duty vs. the same crew off-duty in the lounge vs. a hostile buyer panel). Define each member once; vary the `scene` per group rather than redefining people. There's no fixed vocabulary — write it plainly and the model plays it.
@@ -26,15 +26,11 @@ Open by understanding what they're building. Three common shapes — stay open,
 
 Ask which they're after if it isn't obvious, then proceed.
 
-**Persisting a cast already in play.** When you arrive here from a live session — the user spun up an ad-hoc cast inline and wants to keep it — the personas are already drafted and voiced. Don't re-interrogate: capture them as they've been playing, give the group an `id` and name, ask the memory and default questions, and go straight to the write.
+**Persisting a cast already in play.** When you arrive here from a live session — the user spun up an ad-hoc cast inline and wants to keep it — the personas are already drafted and voiced. Don't re-interrogate: capture them as they've been playing, give the group an `id` and name, ask the default question, and go straight to the write.
 
 ## Editing an existing party
 
-When the user wants to change a party that already exists (retune a member's persona, add someone to a group, swap the default), read the current state first so you change rather than clobber: `uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow` returns the merged `party_members`, `party_groups`, and `default_party`. Show the member or group being touched, capture only the delta with the user, and hand that sparse change to `bmad-customize` — it replaces a `party_members`/`party_groups` entry whose `code`/`id` matches and appends the rest, so an edit is just the changed entry, never a full rewrite.
-
-## Keeping new faces from a session
-
-At the end of a remembered party, the room offers to keep the faces that showed up but aren't in its roster — characters cast from an open-cast scene, or members the user added on the fly. They're already drafted and voiced, so don't re-interrogate: capture each as they played (`code`, `name`, `icon`, a one-line `title`, and a `persona` drawn from how they came across), then add them as `party_members`. For a fixed-roster group, also list their codes in the group's `members` so they return as regulars. For an open-cast room, leave `members` empty — listing any member turns the room into a fixed roster and kills its on-the-fly casting; the saved personas now live in the collective, so the scene still names them and they can return without locking the room down. Hand that sparse delta to `bmad-customize` — for a built-in party with no override yet it creates one; for an existing override it merges the new members in.
+When the user wants to change a party that already exists (retune a member's persona, add someone to a group, swap the default), read the current state first so you change rather than clobber: `python3 {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow` returns the merged `party_members`, `party_groups`, and `default_party`. Show the member or group being touched, capture only the delta with the user, and hand that sparse change to `bmad-customize` — it replaces a `party_members`/`party_groups` entry whose `code`/`id` matches and appends the rest, so an edit is just the changed entry, never a full rewrite.
 
 ## Distill from source data (when provided)
 
@@ -58,13 +54,12 @@ Keep pushing for specificity. "Skeptical CFO" is a placeholder; "won't approve a
 ## Close it out
 
 - Ask straight: **anything else about this party to specify** before you write it — a house dynamic, a missing voice, a member who should lead.
-- Ask whether **this party should remember across sessions** (unless the user already said). Yes → `memory = true` on the group; no → `memory = false`. One-offs with no group skip this — memory is a group setting.
 - Ask whether **this group should be the default party going forward**. Yes → set `default_party` to the group's id. One-offs with no group can't be a default; skip the ask.
 
 ## Write via bmad-customize
 
-**First, check for code collisions.** A custom member whose `code` matches an installed agent silently *overrides* that agent in the collective. Before composing, resolve the collective once — `uv run {skill-root}/scripts/resolve_party.py --project-root {project-root} --skill {skill-root}` — and check each new member's `code` against the returned members. On a collision, surface it ("`analyst` would override the installed Analyst — intended, or pick a different code?") and let the user confirm or rename. One check, not a gate.
+**First, check for code collisions.** A custom member whose `code` matches an installed agent silently *overrides* that agent in the collective. Before composing, resolve the collective once — `python3 {skill-root}/scripts/resolve_party.py --project-root {project-root} --skill {skill-root}` — and check each new member's `code` against the returned members. On a collision, surface it ("`analyst` would override the installed Analyst — intended, or pick a different code?") and let the user confirm or rename. One check, not a gate.
 
-Compose the sparse override and hand it to `bmad-customize` to place, confirm, and write — target skill `bmad-party-mode`, `[workflow]` surface. Default to the **user** override (`bmad-party-mode.user.toml`); offer the **team** file when the party is meant to be shared. Hand it the exact entries: the `party_members` tables, any `party_groups` table (including its `memory` flag), and `default_party` if the user opted in. Keep it sparse — only the new entries, never a copy of the base customize.toml. `bmad-customize` shows the TOML, waits for an explicit yes, writes, and verifies the merge; don't write the file yourself.
+Compose the sparse override and hand it to `bmad-customize` to place, confirm, and write — target skill `bmad-party-mode`, `[workflow]` surface. Default to the **user** override (`bmad-party-mode.user.toml`); offer the **team** file when the party is meant to be shared. Hand it the exact entries: the `party_members` tables, any `party_groups` table, and `default_party` if the user opted in. Keep it sparse — only the new entries, never a copy of the base customize.toml. `bmad-customize` shows the TOML, waits for an explicit yes, writes, and verifies the merge; don't write the file yourself.
 
 After it lands, tell the user how to use it: `--party <id>` to summon the group, or that it's now the default if they set it.

src/core-skills/bmad-party-mode/references/party-memory.md

@@ -1,51 +0,0 @@
-# Party Memory
-
-The room remembers its past sessions with this user and brings them back to life — in character. Memory is per-party and append-only.
-
-Memory is on when the active party's `memory_enabled` is true — the default room follows `{workflow.party_memory}`, a named group its own `memory` flag (both resolved by `resolve_party.py`); ad-hoc inline casts have none. Read on entry and on any mid-session room switch; write through the session.
-
-## Where it lives
-
-One memlog per party: `{workflow.memory_dir}/{active}/.memlog.md`, where `{active}` is the key `resolve_party.py` already returned — the group id (e.g. `code-review-crew`), or `installed` for the default room. The folder is named after the party.
-
-## Read it on entry — distill, don't dump
-
-The log is append-only and grows every session, so don't pull the raw file into the party. Hand a reader subagent the memlog path (`{workflow.memory_dir}/{active}/.memlog.md`) and have it return a compact brief — a few hundred tokens of *where things stand now*, ready to play in character.
-
-Then let the brief shape the room from the first beat, **in character**: behavioral state resumes (a cold pair opens cold, an alliance opens warm), threads pick up, callbacks land when they fit — organically, not recited on sight. Never break the fourth wall: the room *remembers*; it never announces it loaded anything, and forces nothing that doesn't fit.
-
-## When to write
-
-- **When a memorable beat lands** — a clash that shifts the room's temperature, an alliance forming, a line worth a future callback, a decision, an outcome.
-- **A floor.** Once a couple of real exchanges are in from the start, even if nothing dramatic happened, capture what it's about and the opening dynamic.
-
-At wrap-up, if the user does signal done, top up with the final outcome and anything memorable not yet captured.
-
-Writes are silent. The room never announces "noted" or "I'll remember".
-
-## What's worth remembering
-
-The test for every entry: *would this color a future session, or make a callback land, or improve the party?* If not, leave it out. A handful of entries, never a recap, never a transcript. keep each entry as brief as possible but usable by future llm.
-
-## New faces
-
-When a character shows up who isn't in the party's roster — cast from an open-cast scene, or one the user adds on the fly — name them in the entry that captures the moment ("<name> turned up and …") so a recurring face can return next session. At wrap-up these are the faces the room offers to keep, saved into the party's roster through `references/create-party.md` (which writes via `bmad-customize`). Until saved they live only in the memlog, and the room re-conjures them from there.
-
-## Write it
-
-```
-uv run {project-root}/_bmad/scripts/memlog.py append \
-  --workspace {workflow.memory_dir}/{active} \
-  --type <dynamic|moment|callback|outcome> \
-  --text "<one succinct line, in the room's own read of it>"
-```
-
-Add `--by <persona-code>` when a memory belongs to one character. Choose `init` vs `append` from the existence fact you already hold: the entry-read (and, on a mid-session room switch, that room's read) told you whether the memlog exists — `init --workspace {workflow.memory_dir}/{active}` once before the first append when it doesn't, plain `append` when it does. (`init` errors if the file already exists, so don't call it blind.)
-
-If `memlog.py` is unavailable or a write errors, skip it silently and never stall the party on a failed write.
-
-## Forget
-
-The memlog is append-only by design — no surgical delete. To wipe a party's memory, delete its folder (`{workflow.memory_dir}/{active}/`). To correct a wrong memory, append a new entry that supersedes it; the room reads the latest state.
-
-Keep entries sparse. The distilled read keeps the *room* lean no matter how big the log gets, but the on-disk file still grows append-only.

src/core-skills/bmad-party-mode/scripts/resolve_party.py

@@ -197,8 +197,7 @@ def group_detail(g, collective, index):
     raw_members = g.get("members", []) or []
     members, unresolved = resolve_members(raw_members, collective, index)
     detail = {"active": g["id"], "name": g.get("name", g["id"]),
-              "members": members, "unresolved": unresolved,
-              "memory_enabled": bool(g.get("memory", False))}
+              "members": members, "unresolved": unresolved}
     if g.get("scene"):
         detail["scene"] = g["scene"]
     if not raw_members:
@@ -221,9 +220,6 @@ def main():
     groups = workflow.get("party_groups", []) or []
     default_party = workflow.get("default_party", "") or ""
     party_mode = workflow.get("party_mode", "session") or "session"
-    # The global party_memory flag governs only the DEFAULT installed-agent room;
-    # a named group carries its own `memory` flag (resolved in group_detail).
-    party_memory = bool(workflow.get("party_memory", True))
 
     # Group menu never needs the (more expensive) installed-agent resolve.
     if args.list_groups:
@@ -256,8 +252,7 @@ def main():
         # No default group: the installed agents (custom additions stay in the
         # pool but don't crowd the default room), exactly like a plain install.
         result.update({"active": "installed",
-                       "members": [collective[c] for c in installed_codes],
-                       "memory_enabled": party_memory})
+                       "members": [collective[c] for c in installed_codes]})
     _emit(result)
 
 

src/core-skills/bmad-party-mode/scripts/tests/test-resolve_party.py

@@ -113,14 +113,6 @@ class TestGroupDetail(unittest.TestCase):
         self.assertEqual(d["members"], [])
         self.assertEqual(d["scene"][:7], "Figures")
 
-    def test_memory_enabled_follows_group_flag_and_defaults_off(self):
-        on = rp.group_detail({"id": "g", "members": ["morpheus"], "memory": True}, self.col, self.idx)
-        self.assertTrue(on["memory_enabled"])
-        off = rp.group_detail({"id": "g", "members": ["morpheus"], "memory": False}, self.col, self.idx)
-        self.assertFalse(off["memory_enabled"])
-        absent = rp.group_detail({"id": "g", "members": ["morpheus"]}, self.col, self.idx)
-        self.assertFalse(absent["memory_enabled"])  # opt-in per named group
-
 
 class TestInstalledCodesIsDefaultRoom(unittest.TestCase):
     """The default room is installed agents only; pure customs stay in the pool."""

src/core-skills/bmad-spec/SKILL.md

@@ -18,7 +18,7 @@ Multiple skills may call to update the same spec over time.
 
 ## On Activation
 
-1. Resolve customization: `uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly.
+1. Resolve customization: `python3 {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --key workflow`. On failure, read `{skill-root}/customize.toml` directly.
 2. Run `{workflow.activation_steps_prepend}`. Treat `{workflow.persistent_facts}` as foundational context (`file:` entries are loaded).
 3. Load `{project-root}/_bmad/core/config.yaml` (and `config.user.yaml` if present), root level and `bmm` section. Resolve `{user_name}`, `{communication_language}`, `{document_output_language}`, `{planning_artifacts}`, `{project_name}`, `{date}`.
 4. Detect mode. **Headless** when any of: no TTY, programmatic caller (another skill or non-interactive runner), or the first message pre-supplies all inputs and asks for an artifact path back. **Interactive** otherwise. In interactive mode, greet by `{user_name}` in `{communication_language}`, stay in that language, and mention that `bmad-party-mode` and `bmad-advanced-elicitation` are available for deeper exploration on any field.
@@ -57,8 +57,8 @@ Deriving the contract from a living log instead of editing the contract in place
 
 Writes go through the shared script — `{project-root}/_bmad/scripts/memlog.py`, the same location as `resolve_customization.py` (atomic; never read it back except to resume):
 
-- `uv run {project-root}/_bmad/scripts/memlog.py init --workspace {spec-folder} --field topic="<what is being specced>"` — once, at create.
-- `uv run {project-root}/_bmad/scripts/memlog.py append --workspace {spec-folder} --type <decision|constraint|capability|assumption|question|direction|note|event> --text "<one-line gist, reason included>"` — as each lands.
+- `python3 {project-root}/_bmad/scripts/memlog.py init --workspace {spec-folder} --field topic="<what is being specced>"` — once, at create.
+- `python3 {project-root}/_bmad/scripts/memlog.py append --workspace {spec-folder} --type <decision|constraint|capability|assumption|question|direction|note|event> --text "<one-line gist, reason included>"` — as each lands.
 - Terminal moments (a validation verdict, "spec finalized") are `--type event` entries; the memlog carries no status field.
 
 ## The Operation

src/core-skills/bmad-spec/assets/spec-template.md

@@ -20,9 +20,9 @@ Name which (or which combination) applies, who is affected, and the backdrop tha
 
 ## Capabilities
 
-- **CAP-1**
-  - **intent:** {One sentence. "User or system can do X to achieve Y." WHAT, not HOW.}
-  - **success:** {Testable or demonstrable criterion. Something a test or a real demonstration can decide.}
+- id: CAP-1
+  intent: {One sentence. "User or system can do X to achieve Y." WHAT, not HOW.}
+  success: {Testable or demonstrable criterion. Something a test or a real demonstration can decide.}
 
 ## Constraints
 

src/core-skills/module-help.csv

@@ -11,4 +11,3 @@ Core,bmad-review-adversarial-general,Adversarial Review,AR,"Use for quality assu
 Core,bmad-review-edge-case-hunter,Edge Case Hunter Review,ECH,Use alongside adversarial review for orthogonal coverage — method-driven not attitude-driven.,,[path],anytime,,,false,,
 Core,bmad-spec,Spec,SP,"Use to distill any intent input (brief, PRD, transcript, brain dump, design folder, mixed multi-source) into a succinct, no-fluff SPEC.md contract + companions that downstream work derives from. Locks the WHAT before the HOW. Works for software, game design, research, editorial, policy, business, anything intent-bearing. Validation mode also available.",,[path],anytime,,,false,{output_folder}/specs/spec-{slug},SPEC.md + companion files
 Core,bmad-customize,BMad Customize,BC,"Use when you want to change how an agent or workflow behaves — add persistent facts, swap templates, insert activation hooks, or customize menus. Scans what's customizable, picks the right scope (agent vs workflow), writes the override to _bmad/custom/, and verifies the merge. No TOML hand-authoring required.",,,anytime,,,false,{project-root}/_bmad/custom,TOML override files
-Core,bmad-forge-idea,Forge Idea,FI,"Use to pressure-test and harden an idea — software, business, creative, research, or life — until it proves out, hardens into something buildable, or dies cheaply. Persona-driven interrogation; optional handoff to bmad-spec or bmad-quick-dev.",,,anytime,,,false,{output_folder}/forge,refined-idea brief (optional)

src/scripts/resolve_config.py

@@ -10,14 +10,12 @@ Reads from four layers (highest priority last):
 
 Outputs merged JSON to stdout. Errors go to stderr.
 
-Uses only the Python stdlib (`tomllib`) — no third-party dependencies.
-BMad is standardizing on `uv run` to invoke scripts (uv provisions a suitable
-interpreter for you); a plain `python3` on PATH still works during the
-transition. Either runner needs Python 3.11+ for `tomllib`.
+Requires Python 3.11+ (uses stdlib `tomllib`). No `uv`, no `pip install`,
+no virtualenv — plain `python3` is sufficient.
 
-  uv run resolve_config.py --project-root /abs/path/to/project
-  uv run resolve_config.py --project-root ... --key core
-  uv run resolve_config.py --project-root ... --key agents
+  python3 resolve_config.py --project-root /abs/path/to/project
+  python3 resolve_config.py --project-root ... --key core
+  python3 resolve_config.py --project-root ... --key agents
 
 Merge rules (same as resolve_customization.py):
   - Scalars: override wins

src/scripts/resolve_customization.py

@@ -11,14 +11,12 @@ Skill name is derived from the basename of the skill directory.
 
 Outputs merged JSON to stdout. Errors go to stderr.
 
-Uses only the Python stdlib (`tomllib`) — no third-party dependencies.
-BMad is standardizing on `uv run` to invoke scripts (uv provisions a suitable
-interpreter for you); a plain `python3` on PATH still works during the
-transition. Either runner needs Python 3.11+ for `tomllib`.
+Requires Python 3.11+ (uses stdlib `tomllib`). No `uv`, no `pip install`,
+no virtualenv — plain `python3` is sufficient.
 
-  uv run resolve_customization.py --skill /abs/path/to/skill-dir
-  uv run resolve_customization.py --skill ... --key agent
-  uv run resolve_customization.py --skill ... --key agent.menu
+  python3 resolve_customization.py --skill /abs/path/to/skill-dir
+  python3 resolve_customization.py --skill ... --key agent
+  python3 resolve_customization.py --skill ... --key agent.menu
 
 Merge rules (purely structural — no field-name special-casing):
   - Scalars (string, int, bool, float): override wins

test/test-installation-components.js

@@ -3319,63 +3319,134 @@ async function runTests() {
   console.log('');
 
   // ============================================================
-  // Test Suite 46: uv environment check (version parsing + messaging)
+  // Test Suite 46: Python environment check (version parsing + classification)
   // ============================================================
-  console.log(`${colors.yellow}Test Suite 46: uv-check version parsing and messaging${colors.reset}\n`);
+  console.log(`${colors.yellow}Test Suite 46: python-check version parsing and classification${colors.reset}\n`);
 
   try {
-    const { parseUvVersion, detectUv } = require('../tools/installer/core/uv-check');
+    const { parsePythonVersion, classifyPython, detectPython } = require('../tools/installer/core/python-check');
 
     // Version parsing
-    const plain = parseUvVersion('uv 0.5.31');
-    assert(plain && plain.major === 0 && plain.minor === 5 && plain.patch === 31, 'parses "uv 0.5.31"');
-    const brew = parseUvVersion('uv 0.5.31 (Homebrew 2025-02-12)');
-    assert(brew && brew.raw === '0.5.31', 'parses uv version with build suffix');
-    const noPatch = parseUvVersion('uv 1.2');
+    const v312 = parsePythonVersion('Python 3.12.1');
+    assert(v312 && v312.major === 3 && v312.minor === 12 && v312.patch === 1, 'parses "Python 3.12.1"');
+    const v311 = parsePythonVersion('Python 3.11.0\n');
+    assert(v311 && v311.raw === '3.11.0', 'parses with trailing newline');
+    const v2 = parsePythonVersion('\nPython 2.7.18');
+    assert(v2 && v2.major === 2, 'parses Python 2 output (stderr-style)');
+    const noPatch = parsePythonVersion('Python 3.13');
     assert(noPatch && noPatch.patch === 0, 'missing patch defaults to 0');
-    assert(parseUvVersion('') === null, 'empty output returns null');
-    assert(parseUvVersion('command not found: uv') === null, 'non-version output returns null');
-    assert(parseUvVersion(null) === null, 'null output returns null');
+    assert(parsePythonVersion('') === null, 'empty output returns null');
+    assert(parsePythonVersion('command not found: python3') === null, 'non-version output returns null');
+    assert(parsePythonVersion(null) === null, 'null output returns null');
 
-    // Detection smoke test — must not throw; result is null or well-formed.
-    const detectedUv = detectUv();
-    assert(detectedUv === null || typeof detectedUv.version.raw === 'string', 'detectUv returns null or a well-formed result');
+    // Classification against feature requirements
+    assert(classifyPython({ major: 3, minor: 11 }) === 'full', '3.11 is full support (tomllib floor)');
+    assert(classifyPython({ major: 3, minor: 13 }) === 'full', '3.13 is full support');
+    assert(classifyPython({ major: 4, minor: 0 }) === 'full', 'hypothetical 4.0 is full support');
+    assert(classifyPython({ major: 3, minor: 10 }) === 'partial', '3.10 is partial (memlog yes, tomllib no)');
+    assert(classifyPython({ major: 3, minor: 8 }) === 'partial', '3.8 is partial (memlog floor)');
+    assert(classifyPython({ major: 3, minor: 7 }) === 'unsupported', '3.7 is unsupported');
+    assert(classifyPython({ major: 2, minor: 7 }) === 'unsupported', '2.7 is unsupported');
+    assert(classifyPython(null) === 'none', 'no python is none');
 
-    // checkUvEnvironment branch coverage — stub detection + prompts so the
-    // assertions are deterministic regardless of whether uv is installed.
-    const uvCheck = require('../tools/installer/core/uv-check');
+    // Detection smoke test — must not throw, and if it finds a Python the
+    // result must be well-formed. (CI machines may or may not have Python.)
+    const detected = detectPython();
+    assert(
+      detected === null ||
+        (typeof detected.command === 'string' &&
+          typeof detected.version.raw === 'string' &&
+          typeof detected.isRuntimeCommand === 'boolean'),
+      'detectPython returns null or a well-formed result',
+    );
+
+    // checkPythonEnvironment branch coverage — stub detection, prompts, and
+    // process.exit so the assertions are deterministic regardless of the
+    // machine's Python. python-check resolves detectPython via module.exports
+    // and prompts via the shared module object, so swapping properties works.
+    const pythonCheck = require('../tools/installer/core/python-check');
     const promptsModule = require('../tools/installer/prompts');
-    const realUv = { detectUv: uvCheck.detectUv, log: promptsModule.log, note: promptsModule.note };
-    const stubUv = (detectResult) => {
-      const seen = { success: [], warn: [], note: [] };
-      uvCheck.detectUv = () => detectResult;
+    const real = {
+      detectPython: pythonCheck.detectPython,
+      log: promptsModule.log,
+      note: promptsModule.note,
+      select: promptsModule.select,
+      cancel: promptsModule.cancel,
+      exit: process.exit,
+    };
+    const stub = (detectResult, selectAnswer) => {
+      const seen = { success: [], warn: [], info: [], note: [], select: [], cancel: [], exit: [] };
+      pythonCheck.detectPython = () => detectResult;
       promptsModule.log = {
         success: async (m) => void seen.success.push(m),
         warn: async (m) => void seen.warn.push(m),
-        info: async () => {},
+        info: async (m) => void seen.info.push(m),
         error: async () => {},
       };
       promptsModule.note = async (m, t) => void seen.note.push(t || m);
+      promptsModule.select = async (opts) => {
+        seen.select.push(opts.message);
+        return selectAnswer;
+      };
+      promptsModule.cancel = async (m) => void seen.cancel.push(m);
+      process.exit = (code) => {
+        seen.exit.push(code);
+        throw new Error('__stub_exit__');
+      };
       return seen;
     };
 
     try {
-      // Branch: uv present — success, no warning.
-      let seen = stubUv({ version: { major: 0, minor: 5, patch: 31, raw: '0.5.31' } });
-      let result = await uvCheck.checkUvEnvironment();
-      assert(result.status === 'found' && seen.success.length === 1, 'uv present logs success');
-      assert(seen.success[0].includes('uv run') && seen.warn.length === 0, 'uv present mentions uv run, no warning');
+      const v = (major, minor, patch) => ({ major, minor, patch, raw: `${major}.${minor}.${patch}` });
 
-      // Branch: uv missing — warn + setup note, never blocks (no prompt).
-      seen = stubUv(null);
-      result = await uvCheck.checkUvEnvironment();
-      assert(result.status === 'missing' && seen.warn.length === 1, 'uv missing warns');
-      assert(seen.warn[0].includes('de facto standard'), 'uv-missing warning frames uv as the de facto standard');
-      assert(seen.note.length === 1 && seen.note[0].includes('uv'), 'uv missing shows a setup note');
+      // Branch: full support via the runtime command — success, no prompt.
+      let seen = stub({ command: 'python3', version: v(3, 12, 1), isRuntimeCommand: true }, 'continue');
+      let result = await pythonCheck.checkPythonEnvironment();
+      assert(result.status === 'full' && seen.success.length === 1, 'full support via python3 logs success');
+      assert(seen.select.length === 0 && seen.warn.length === 0, 'full support via python3 skips warning and ack prompt');
+
+      // Branch: modern Python found, but not as `python3` — runtime mismatch.
+      seen = stub({ command: 'py -3', version: v(3, 12, 0), isRuntimeCommand: false }, 'continue');
+      result = await pythonCheck.checkPythonEnvironment();
+      assert(seen.success.length === 0, 'python3-mismatch never reports full support');
+      assert(
+        seen.warn.length === 1 && seen.warn[0].includes('python3') && seen.warn[0].includes('py -3'),
+        'python3-mismatch warns that scripts invoke python3',
+      );
+      assert(seen.select.length === 1 && result.status === 'full', 'python3-mismatch still requires the ack prompt');
+
+      // Branch: partial support (3.8–3.10) — warn + ack, continue returns.
+      seen = stub({ command: 'python3', version: v(3, 9, 5), isRuntimeCommand: true }, 'continue');
+      result = await pythonCheck.checkPythonEnvironment();
+      assert(
+        result.status === 'partial' && seen.warn.length === 1 && seen.warn[0].includes('3.11+'),
+        'partial support warns about tomllib floor',
+      );
+      assert(seen.select.length === 1 && seen.exit.length === 0, 'partial support prompts and continue proceeds');
+
+      // Branch: no Python, non-interactive — warn + info, never prompts.
+      seen = stub(null, 'continue');
+      result = await pythonCheck.checkPythonEnvironment({ nonInteractive: true });
+      assert(result.status === 'none' && seen.warn[0].includes('No Python found'), 'non-interactive with no Python warns');
+      assert(seen.select.length === 0 && seen.info.length === 1, 'non-interactive skips the ack prompt and logs continuation');
+
+      // Branch: no Python, interactive, user quits — cancel message + exit 0.
+      seen = stub(null, 'quit');
+      let threw = false;
+      try {
+        await pythonCheck.checkPythonEnvironment();
+      } catch (error) {
+        threw = error.message === '__stub_exit__';
+      }
+      assert(threw && seen.exit.length === 1 && seen.exit[0] === 0, 'quit choice exits 0 (user-cancel convention)');
+      assert(seen.cancel.length === 1, 'quit choice shows the cancel guidance');
     } finally {
-      uvCheck.detectUv = realUv.detectUv;
-      promptsModule.log = realUv.log;
-      promptsModule.note = realUv.note;
+      pythonCheck.detectPython = real.detectPython;
+      promptsModule.log = real.log;
+      promptsModule.note = real.note;
+      promptsModule.select = real.select;
+      promptsModule.cancel = real.cancel;
+      process.exit = real.exit;
     }
   } catch (error) {
     console.log(`${colors.red}Test Suite 46 setup failed: ${error.message}${colors.reset}`);

tools/installer/core/installer.js

@@ -1233,9 +1233,6 @@ class Installer {
       `    1. Launch your AI agent from your project folder`,
       `    2. Not sure what to do? Invoke the ${color.cyan('bmad-help')} skill and ask it what to do!`,
       '',
-      `    ${color.cyan('Tip:')} BMAD workflows increasingly run Python scripts via ${color.cyan('uv run')} — uv is`,
-      `    becoming the de facto standard. If you don't have it yet, ask your agent to set it up.`,
-      '',
       `    Blog, Docs and Guides: ${color.blue('https://bmadcode.com/')}`,
       `    Community: ${color.blue('https://discord.gg/gk8jAdXWmj')}`,
     );

tools/installer/core/python-check.js

@@ -0,0 +1,199 @@
+const { spawnSync } = require('node:child_process');
+const prompts = require('../prompts');
+
+// Python 3.11 added stdlib `tomllib` (PEP 680), which the shared scripts in
+// src/scripts/ (resolve_config.py, resolve_customization.py) require to read
+// BMAD's TOML config files. memlog.py is more lenient and runs on 3.8+.
+const PYTHON_FULL_SUPPORT = { major: 3, minor: 11 };
+const PYTHON_PARTIAL_SUPPORT = { major: 3, minor: 8 };
+
+// Every runtime call site (skill steps, on_complete hooks) invokes a literal
+// `python3`, so only that command's version vouches for BMAD features. The
+// fallback probes exist to tell the user "Python is installed, but not under
+// the name BMAD uses" instead of a misleading "No Python found".
+const RUNTIME_COMMAND = 'python3';
+const PROBE_CANDIDATES =
+  process.platform === 'win32'
+    ? [
+        { command: 'python3', args: ['--version'] },
+        { command: 'py', args: ['-3', '--version'] },
+        { command: 'python', args: ['--version'] },
+      ]
+    : [
+        { command: 'python3', args: ['--version'] },
+        { command: 'python', args: ['--version'] },
+      ];
+
+/**
+ * Parse a `python --version` output line into version parts.
+ * Python 3 prints to stdout; Python 2 printed to stderr — callers pass both.
+ * @param {string} output - Combined stdout/stderr from `python --version`
+ * @returns {{major: number, minor: number, patch: number, raw: string}|null}
+ */
+function parsePythonVersion(output) {
+  if (!output) return null;
+  const match = output.match(/Python\s+(\d+)\.(\d+)(?:\.(\d+))?/);
+  if (!match) return null;
+  return {
+    major: Number(match[1]),
+    minor: Number(match[2]),
+    patch: Number(match[3] || 0),
+    raw: `${match[1]}.${match[2]}.${match[3] || 0}`,
+  };
+}
+
+/**
+ * Classify a detected Python version against BMAD's feature requirements.
+ * @param {{major: number, minor: number}|null} version
+ * @returns {'full'|'partial'|'unsupported'|'none'}
+ */
+function classifyPython(version) {
+  if (!version) return 'none';
+  const { major, minor } = version;
+  if (major > PYTHON_FULL_SUPPORT.major || (major === PYTHON_FULL_SUPPORT.major && minor >= PYTHON_FULL_SUPPORT.minor)) {
+    return 'full';
+  }
+  if (major === PYTHON_PARTIAL_SUPPORT.major && minor >= PYTHON_PARTIAL_SUPPORT.minor) {
+    return 'partial';
+  }
+  return 'unsupported';
+}
+
+/**
+ * Run one probe candidate and return its parsed version, or null.
+ * @param {{command: string, args: string[]}} candidate
+ * @returns {{major: number, minor: number, patch: number, raw: string}|null}
+ */
+function probeVersion(candidate) {
+  const run = (extra = {}) =>
+    spawnSync(candidate.command, candidate.args, {
+      encoding: 'utf8',
+      timeout: 5000,
+      windowsHide: true,
+      ...extra,
+    });
+  let result = run();
+  // Node >=18.20/20.12 refuses to spawn .bat/.cmd without a shell
+  // (CVE-2024-27980 hardening) and reports EINVAL — pyenv-win ships its
+  // python shims as .bat. Args here are static literals, so a shell retry
+  // is injection-safe.
+  if (result.error && result.error.code === 'EINVAL' && process.platform === 'win32') {
+    result = run({ shell: true });
+  }
+  if (result.error) return null;
+  return parsePythonVersion(`${result.stdout || ''}\n${result.stderr || ''}`);
+}
+
+/**
+ * Probe the local environment for a Python interpreter.
+ * Tries each candidate command and returns the first that reports a version.
+ * `isRuntimeCommand` is true only when the match is `python3` — the command
+ * BMAD scripts actually invoke.
+ * @returns {{command: string, version: {major: number, minor: number, patch: number, raw: string}, isRuntimeCommand: boolean}|null}
+ */
+function detectPython() {
+  for (const candidate of PROBE_CANDIDATES) {
+    try {
+      const version = probeVersion(candidate);
+      if (version) {
+        const display = candidate.args.length > 1 ? `${candidate.command} ${candidate.args.slice(0, -1).join(' ')}` : candidate.command;
+        return { command: display, version, isRuntimeCommand: candidate.command === RUNTIME_COMMAND };
+      }
+    } catch {
+      // Candidate not runnable — try the next one.
+    }
+  }
+  return null;
+}
+
+function upgradeHints() {
+  return [
+    'How to get Python 3.11+ (as `python3`):',
+    '  macOS:        brew install python3',
+    '  Windows:      winget install Python.Python.3.12  (then ensure `python3` resolves, e.g. enable the python3 alias)',
+    '  Linux/WSL:    sudo apt install python3  (Ubuntu 24.04+ ships 3.12; older distros: use pyenv or deadsnakes)',
+    '  Docker:       add python3 to your image (e.g. apk add python3 / apt-get install -y python3)',
+  ].join('\n');
+}
+
+/**
+ * Check the local Python environment and warn about degraded BMAD features.
+ *
+ * Warn-don't-block: most of BMAD works without Python, so the install always
+ * may proceed — but the user must explicitly acknowledge the warning so it
+ * can't scroll past unseen. In non-interactive runs (--yes, or stdin is not
+ * a TTY) the warning is logged and the install continues without a prompt.
+ *
+ * @param {Object} [options]
+ * @param {boolean} [options.nonInteractive=false] - Skip the ack prompt (--yes, or no TTY)
+ * @returns {Promise<{status: string, detected: Object|null}>}
+ */
+async function checkPythonEnvironment({ nonInteractive = false } = {}) {
+  // Called via module.exports so tests can stub detection.
+  const detected = module.exports.detectPython();
+  const status = classifyPython(detected ? detected.version : null);
+
+  if (status === 'full' && detected.isRuntimeCommand) {
+    await prompts.log.success(`Python ${detected.version.raw} detected (${detected.command}) — all BMAD features supported.`);
+    return { status, detected };
+  }
+
+  if (detected && !detected.isRuntimeCommand) {
+    await prompts.log.warn(
+      `Python ${detected.version.raw} found via \`${detected.command}\`, but BMAD scripts invoke \`python3\`, which is not on PATH.\n` +
+        `Python-powered features (memlog session memory, TOML config resolution) won't run until \`python3\` resolves —\n` +
+        `add a python3 alias/shim, or reinstall Python with the python3 launcher enabled.`,
+    );
+  } else if (status === 'partial') {
+    await prompts.log.warn(
+      `Python ${detected.version.raw} detected (${detected.command}) — BMAD's TOML config tools need Python 3.11+ (stdlib tomllib).\n` +
+        `Works: memlog session memory. Won't work: config/customization resolution scripts.`,
+    );
+  } else {
+    const found =
+      status === 'unsupported' ? `Python ${detected.version.raw} detected (${detected.command}) — too old.` : 'No Python found on PATH.';
+    await prompts.log.warn(
+      `${found} BMAD installs fine without it, but Python-powered features\n` +
+        `(memlog session memory, TOML config resolution) won't run until Python 3.11+ is available.`,
+    );
+  }
+  await prompts.note(upgradeHints(), 'Python 3.11+ recommended');
+
+  if (nonInteractive) {
+    await prompts.log.info('Continuing anyway (non-interactive run). You can fix Python later — no reinstall needed.');
+    return { status, detected };
+  }
+
+  const choice = await prompts.select({
+    message: "BMAD's Python-powered features won't work yet. How do you want to proceed?",
+    choices: [
+      {
+        name: 'Continue install',
+        value: 'continue',
+        hint: 'BMAD works without Python — you can fix Python later, no reinstall needed',
+      },
+      {
+        name: 'Quit and fix Python first',
+        value: 'quit',
+        hint: 'make Python 3.11+ available as python3, then re-run the installer',
+      },
+    ],
+    default: 'continue',
+  });
+
+  if (choice === 'quit') {
+    await prompts.cancel('Make Python 3.11+ available as `python3` (see hints above), then re-run the installer.');
+    process.exit(0);
+  }
+
+  return { status, detected };
+}
+
+module.exports = {
+  checkPythonEnvironment,
+  detectPython,
+  parsePythonVersion,
+  classifyPython,
+  PYTHON_FULL_SUPPORT,
+  PYTHON_PARTIAL_SUPPORT,
+};

tools/installer/core/uv-check.js

@@ -1,97 +0,0 @@
-const { spawnSync } = require('node:child_process');
-const prompts = require('../prompts');
-
-// `uv` (https://docs.astral.sh/uv/) is becoming the de facto standard for
-// running the Python scripts BMAD workflows shell out to: `uv run <script>`
-// resolves the interpreter and any dependencies on demand, so skills don't
-// have to assume a particular `python3` is on PATH. The ecosystem is mid-
-// migration — some skills still call `python3` directly — so a missing `uv`
-// is a warning, not a blocker: BMAD installs and runs either way.
-const RUNTIME_COMMAND = 'uv';
-
-/**
- * Parse `uv --version` output into version parts.
- * Example outputs: "uv 0.5.31", "uv 0.5.31 (Homebrew 2025-02-12)".
- * @param {string} output - stdout/stderr from `uv --version`
- * @returns {{major: number, minor: number, patch: number, raw: string}|null}
- */
-function parseUvVersion(output) {
-  if (!output) return null;
-  const match = output.match(/uv\s+(\d+)\.(\d+)(?:\.(\d+))?/i);
-  if (!match) return null;
-  return {
-    major: Number(match[1]),
-    minor: Number(match[2]),
-    patch: Number(match[3] || 0),
-    raw: `${match[1]}.${match[2]}.${match[3] || 0}`,
-  };
-}
-
-/**
- * Probe the local environment for `uv`.
- * @returns {{version: {major: number, minor: number, patch: number, raw: string}}|null}
- */
-function detectUv() {
-  let result;
-  try {
-    result = spawnSync(RUNTIME_COMMAND, ['--version'], {
-      encoding: 'utf8',
-      timeout: 5000,
-      windowsHide: true,
-    });
-  } catch {
-    return null;
-  }
-  if (!result || result.error) return null;
-  const version = parseUvVersion(`${result.stdout || ''}\n${result.stderr || ''}`);
-  return version ? { version } : null;
-}
-
-function setupHints() {
-  return [
-    'BMAD workflows increasingly run Python scripts via `uv run`, which manages',
-    'the interpreter and dependencies for you — no manual venv or pip needed.',
-    '',
-    'Easiest path: ask your AI agent to "install and set up uv for me".',
-    '',
-    'Or install it yourself:',
-    '  macOS/Linux:  curl -LsSf https://astral.sh/uv/install.sh | sh',
-    '  Windows:      powershell -c "irm https://astral.sh/uv/install.ps1 | iex"',
-    '  Homebrew:     brew install uv',
-    '  Docs:         https://docs.astral.sh/uv/getting-started/installation/',
-  ].join('\n');
-}
-
-/**
- * Check whether `uv` is available and inform the user.
- *
- * Warn-don't-block, and no acknowledgement prompt: `uv` is on its way to being
- * the standard runner for BMAD's Python scripts, but the migration is still in
- * progress, so the install never stops on its account. The note tells the user
- * how to set it up (preferably by asking their agent).
- *
- * @returns {Promise<{status: 'found'|'missing', detected: Object|null}>}
- */
-async function checkUvEnvironment() {
-  // Called via module.exports so tests can stub detection.
-  const detected = module.exports.detectUv();
-
-  if (detected) {
-    await prompts.log.success(`uv ${detected.version.raw} detected — ready to run BMAD's Python-powered scripts via \`uv run\`.`);
-    return { status: 'found', detected };
-  }
-
-  await prompts.log.warn(
-    "uv not found on PATH. uv is becoming the de facto standard for running BMAD's Python\n" +
-      'scripts (`uv run <script>`), and it provisions the interpreter for you. BMAD installs\n' +
-      'fine without it, but setting up uv now keeps you ahead as workflows adopt it.',
-  );
-  await prompts.note(setupHints(), 'uv recommended');
-  return { status: 'missing', detected: null };
-}
-
-module.exports = {
-  checkUvEnvironment,
-  detectUv,
-  parseUvVersion,
-};

tools/installer/ide/platform-codes.yaml

@@ -162,13 +162,6 @@ platforms:
       target_dir: .agents/skills
       global_target_dir: ~/.config/agents/skills
 
-  hermes:
-    name: "Hermes Agent"
-    preferred: false
-    installer:
-      target_dir: .agents/skills
-      global_target_dir: ~/.hermes/skills
-
   iflow:
     name: "iFlow"
     preferred: false

tools/installer/install-messages.yaml

@@ -12,10 +12,6 @@ startMessage: |
   🌟 100% free. 100% open source. Always.
      No paywalls. No gated content. Knowledge shared, not sold.
 
-  🐍 HEADS UP: uv (https://docs.astral.sh/uv/) is becoming the de facto standard
-     for running the Python scripts BMAD workflows rely on (`uv run <script>`).
-     If it's not set up yet, ask your AI agent to "install and set up uv for me".
-
   🌐 CONNECT:
     Website:   https://bmadcode.com/
     Discord:   https://discord.gg/gk8jAdXWmj

tools/installer/ui.js

@@ -161,16 +161,15 @@ class UI {
     const messageLoader = new MessageLoader();
     await messageLoader.displayStartMessage();
 
-    // Probe for `uv` before any other prompts: it's becoming the de facto
-    // runner for the Python scripts BMAD workflows shell out to
-    // (`uv run <script>`), and uv provisions the interpreter itself, so it's
-    // the single thing worth checking for. The migration is still in progress
-    // (some skills still call `python3` directly), so this is informational —
-    // warn-don't-block, no ack prompt — and just points the user at setup
-    // (ideally "ask your agent to set up uv"). The installer runs in the
-    // destination environment, so probing PATH here tests the right machine.
-    const { checkUvEnvironment } = require('./core/uv-check');
-    await checkUvEnvironment();
+    // Probe the local Python before any other prompts: several BMAD features
+    // (memlog session memory, TOML config resolution) need Python 3.11+ at
+    // runtime. Warn-don't-block, but require an explicit ack so the warning
+    // can't scroll past unseen. The installer runs in the destination
+    // environment, so probing PATH here tests the right machine.
+    // Skip the ack when stdin isn't a TTY (CI/Docker/piped): clack's select
+    // on closed stdin resolves to cancel, which would silently exit 0.
+    const { checkPythonEnvironment } = require('./core/python-check');
+    await checkPythonEnvironment({ nonInteractive: !!options.yes || !process.stdin.isTTY });
 
     // Parse channel flags (--channel/--all-*/--next=/--pin) once. Warnings
     // are surfaced immediately so the user sees them before any git ops run.

website/astro.config.mjs

@@ -106,25 +106,25 @@ export default defineConfig({
           label: 'Tutorials',
           translations: { 'vi-VN': 'Hướng dẫn nhập môn', 'zh-CN': '教程', 'fr-FR': 'Tutoriels', 'cs-CZ': 'Tutoriály' },
           collapsed: false,
-          items: [{ autogenerate: { directory: 'tutorials' } }],
+          autogenerate: { directory: 'tutorials' },
         },
         {
           label: 'How-To Guides',
           translations: { 'vi-VN': 'Hướng dẫn tác vụ', 'zh-CN': '操作指南', 'fr-FR': 'Guides pratiques', 'cs-CZ': 'Praktické návody' },
           collapsed: true,
-          items: [{ autogenerate: { directory: 'how-to' } }],
+          autogenerate: { directory: 'how-to' },
         },
         {
           label: 'Explanation',
           translations: { 'vi-VN': 'Giải thích', 'zh-CN': '概念说明', 'fr-FR': 'Explications', 'cs-CZ': 'Vysvětlení' },
           collapsed: true,
-          items: [{ autogenerate: { directory: 'explanation' } }],
+          autogenerate: { directory: 'explanation' },
         },
         {
           label: 'Reference',
           translations: { 'vi-VN': 'Tham chiếu', 'zh-CN': '参考', 'fr-FR': 'Référence', 'cs-CZ': 'Reference' },
           collapsed: true,
-          items: [{ autogenerate: { directory: 'reference' } }],
+          autogenerate: { directory: 'reference' },
         },
         // TEA docs moved to standalone module site; keep BMM sidebar focused.
         {

website/src/content.config.ts

@@ -1,8 +0,0 @@
-import { defineCollection } from 'astro:content';
-import { docsLoader, i18nLoader } from '@astrojs/starlight/loaders';
-import { docsSchema, i18nSchema } from '@astrojs/starlight/schema';
-
-export const collections = {
-  docs: defineCollection({ loader: docsLoader(), schema: docsSchema() }),
-  i18n: defineCollection({ loader: i18nLoader(), schema: i18nSchema() }),
-};

website/src/content/config.ts

@@ -0,0 +1,7 @@
+import { defineCollection } from 'astro:content';
+import { docsSchema, i18nSchema } from '@astrojs/starlight/schema';
+
+export const collections = {
+  docs: defineCollection({ schema: docsSchema() }),
+  i18n: defineCollection({ type: 'data', schema: i18nSchema() }),
+};

website/src/pages/404.astro

@@ -1,10 +1,10 @@
 ---
 import StarlightPage from '@astrojs/starlight/components/StarlightPage.astro';
-import { getEntry, render } from 'astro:content';
+import { getEntry } from 'astro:content';
 import { translatedLocales } from '../lib/locales.mjs';
 
 const entry = await getEntry('docs', '404');
-const { Content } = await render(entry);
+const { Content } = await entry.render();
 ---
 
 <StarlightPage frontmatter={{ title: entry.data.title, template: entry.data.template }}>