Retrospective Step 12 appends agreed action items to a new action_items
section in sprint-status.yaml and updates the previous epic's entries from
the Step 4 follow-through. Sprint-status parses the section, validates its
statuses, and surfaces open items in the summary and data mode.
Sprint-planning preserves the section when regenerating the file.
* bmad-architecture: breadth coverage + lean directives; reviewer reports to subfolder
- Inline forward-readiness (inherit upstream silently; thin input -> suggest bmad-spec or hybrid-capture) and brownfield (ratify, don't re-tell) directives in place of a standalone spine-checklist
- Use {workflow.persistent_facts} instead of hardcoded project-context.md
- Reviewer gate writes per-reviewer reports to reviews/ subfolder so the deliverable folder stays clean
- Require breadth coverage at distill and in the gate rubric: every altitude-owned dimension decided/deferred/open, flagging the operational/environmental envelope a domain-focused draft skips
- Trim repeated directives and human-facing justification prose
* Redesign spine template and move stack pinning to a body table
Rework spine-template.md so it stops forcing fixed structure: guidance
moves into single-line HTML comments (stripped at distill), the always-two
diagrams and empty-mermaid render bugs are gone, and the structural-seed
framing opens up so the operational/environmental envelope isn't skipped.
Stack moves from nested frontmatter into a ## Stack | Name | Version | table.
lint_spine.py drops the frontmatter dep check for find_unpinned_stack, which
parses the Stack table and flags real-name/blank-version rows while skipping
{token} skeletons. Tests reworked to match; 24 passing.
SKILL.md Finalize tightened to act-then-strip template comments and sweep
altitude-owned breadth.
* Harden find_unpinned_stack: blank fences, locate columns, looser heading
Address PR review (CodeRabbit + Augment): find_unpinned_stack scanned raw
body, so pipe-rows or ## headings inside a fenced block could be misread as
live Stack content and misreport version_pin. Now blanks fences first, like
find_placeholders and find_ad_issues, honoring the linter's fences-are-non-live
contract.
Also locate both Name and Version columns from the header (a reordered table
now pairs name to version correctly) and match the heading on a word boundary
(## Stack & Versions still counts). Add regression tests for fenced rows,
fenced headings, renamed heading, and reordered columns (28 passing).
Reword stale 'unpinned deps' / 'unpinned dependency versions' to 'unpinned
Stack versions' to match the body-table model.
* Rework solutioning architecture skill into bmad-architecture (spine)
- Rename canonical skill bmad-tech-plan -> bmad-architecture; output is ARCHITECTURE-SPINE.md
- Convert bmad-create-architecture to a deprecated husk forwarding to bmad-architecture (create intent); strip steps/data/template
- Finalize: spine is the default, then offers fuller renderings + a self-contained HTML view; doc_standards polish applies to prose docs only
- Reviewer Gate: add adversarial divergence-hunter, on by default at high/regulated/cross-team stakes
- Fix references in module-help.csv, bmad-prd, bmad-ux, bmad-agent-architect; reword help entry for when-to-offer
- bmad-spec: flag recognized-but-unaddressed domain implications as open_questions
* bmad-architecture: extract Finalize/reviewer-menu to references, add grade_spine script
- Move the Finalize sequence out of SKILL.md into references/finalize.md
- Make references/validate.md own the canonical reviewer menu and prompts
- Add scripts/grade_spine.py (+ tests) so validation grade is computed
deterministically instead of derived by hand
- Tighten SKILL.md spine/memlog framing
* bmad-architecture: outcome-driven rewrite + spawned reviewer gate
Re-express SKILL.md as goals/outcomes rather than procedure: intent is read
from the input (raw idea, large doc, codebase, feature slice, existing spine)
instead of a Create/Update/Validate x mode matrix. Restore the counter-default
coaching invariants that over-compression had stripped — Guided is the default,
load-bearing calls are shown with alternatives and the user chooses, recommend a
known starter when the stack is open, investigate brownfield before deciding.
Adopt the bmad-prd/product-brief shape: a dedicated Reviewer Gate that always
spawns finalize_reviewers as parallel subagents against the spine (lint floor
first; ad-hoc lenses scaled to rigor/altitude/criticality), and a numbered
Finalize that calls it. Headless runs the full gate non-interactively.
Reframe the structural seed as the living source of truth for shape (code owns
detail; evolve on shape change; memlog keeps history). Fix template mermaid
(C4 -> flowchart, valid one-attr-per-line erDiagram). Ship two default
finalize_reviewers (currency/reality check, adversarial divergence hunter).
Remove grade_spine.py and the inlined discovery/inputs/validate/finalize refs.
* bmad-architecture: review fixes — shared memlog, epic altitude, lint hardening
Pre-PR review fixes for the new architecture-spine skill:
- Adopt the shared canonical memlog (#2462); drop the vendored copy and the
status-flag lifecycle in favor of `event` entries
- Wire epic-altitude inheritance: parent-spine load+bind, Inherited Invariants
template section, epic-scoped run folder, parent-contradiction lens; state
that per-story detail is deferred to bmad-create-story
- Add the Update intent; honor a forwarded-activation handoff from the
deprecated bmad-create-architecture shim
- Rename modes to Coaching path / Fast path (suite-consistent with prd/brief)
and sequence the offer as an activation step
- Fix the brownfield project-context.md default path ({output_folder} no-op)
- lint_spine: line-exact frontmatter, fence-blanking (AD-in-fence + accurate
line numbers), map-form key_deps, AD-0 fix; soften template-token severity;
add 8 tests
- Carve the Reviewer Gate to references/ to stay under the SKILL.md token budget
- Template: entities-only ERD, scale-down guidance, softened seed framing;
gitignore .memlog.md
* bmad-architecture: address PR review (augment + coderabbit)
- REF-03: use canonical "invoke the `skill` skill" language for all cross-skill
references in SKILL.md (was "route to / offer / hand to / Next:")
- lint_spine: scan frontmatter for unfilled template tokens / TBD (paradigm,
scope, date were uncatchable in the body-only pass)
- lint_spine: guard read_text() so a bad-encoding/unreadable spine returns error
JSON and exit 0, honoring the documented contract
- reviewer-gate: resolve the "always run" vs "may skip" ambiguity — the gate
scales/skips by stakes, but finalize_reviewers always run once it does
- tests: +3 (frontmatter token, frontmatter TBD, unreadable spine); use next()
* bmad-architecture: align finalize-reviewers wording + headless blocked output
- customize.toml: reword finalize_reviewers comment to match reviewer-gate.md —
the gate is stakes-gated, but its reviewers always run once it does (coderabbit)
- headless.md: state explicitly that a blocked run omits spine/memlog/companions,
matching the "omit keys for artifacts not produced" contract (coderabbit)
* bmad-architecture: ask deliverable purpose up front, offer presentable renderings, lead next-steps with bmad-spec
* feat: installer detects Python version and warns when 3.11+ (tomllib) is missing
Several BMAD features need Python at runtime: memlog (3.8+) and the TOML
config resolution scripts (3.11+ for stdlib tomllib). Users install into
varied environments (Linux, Windows, WSL, Docker) where Python may be
missing or too old, and previously only found out via runtime errors.
The installer now probes PATH at startup (py -3 / python3 / python) and
classifies the result: 3.11+ passes silently with a success line; 3.8-3.10
or missing/too-old Python gets a warning naming exactly which features
degrade, plus per-platform install hints. The warning requires an explicit
ack — continue (fix later, no reinstall needed) or quit and re-run after
installing Python. Warn-don't-block: most of BMAD works without Python, so
the install is never refused. In --yes mode the warning logs and continues
without prompting.
* fix: align Python check with runtime truth (python3) and harden edge cases
Review fixes for the installer Python check:
- Probe python3 first on all platforms: every runtime call site invokes a
literal python3, so only that command vouches for BMAD features. Python
found via py/python now gets an explicit mismatch warning instead of a
false "all BMAD features supported".
- Treat closed/piped stdin as non-interactive (in addition to --yes) so
scripted installs no longer silently exit 0 via clack's cancel path.
- Retry probes with shell:true on win32 EINVAL (CVE-2024-27980 hardening
rejects .bat/.cmd shims like pyenv-win's without a shell).
- Add Suite 46 branch tests for checkPythonEnvironment with stubbed
detection, prompts, and process.exit.
The gate ported from #2398 defended against runtime customization
indirection: agents guessed resolver outputs instead of executing them,
silently skipping append steps. render.py inlines the prepend/append
entries into the rendered workflow.md, so there is nothing left to
short-circuit, and each inlined list already carries its own execute-
in-order imperative. In the default install both lists render as
_None._ and the gate is pure noise.
* fix: remove empty skill-group dirs left in _bmad after install
Skill cleanup removed each skill's own directory but never pruned the
now-empty grouping folders above it (e.g. _bmad/bmm/1-analysis), leaving
empty dirs behind after every install. Walk up from each removed skill
dir and drop empty parents, stopping at the bmad root.
* fix: harden empty-parent pruning boundary and cleanup
Use a path-boundary check instead of a string prefix so sibling dirs
(e.g. _bmad2) can't match the bmad root, and make the walk best-effort
so a dir that vanishes or fills in mid-walk never aborts the install.
Move the test fixture cleanup into finally so failures don't leak temp
dirs.
---------
Co-authored-by: Brian <bmadcode@gmail.com>
* bmad-spec: make the memlog canonical, SPEC.md a derived view
Replace the bespoke .decision-log.md with the shared memlog script
(_bmad/scripts/memlog.py, same location as resolve_customization.py).
The append-only memlog becomes the single source of truth; SPEC.md and
spec-authored companions are re-derived from it (plus cited sources for
raw content) on each run instead of hand-patched. This makes bmad-spec
the sole writer of the spec and lets the surrounding steps (PRD, UX,
architecture, epics) feed one spec in any order without merge drift.
- New "Memory and derivation" section: memlog canonical, SPEC.md a
projection, single-writer rule, append/init via the shared script,
no status field (terminal moments are event entries).
- Operation reads the prior memlog (not the rendered SPEC.md) as the
authority on decisions and capability IDs on update.
- Conflict-surfacing: live sources/companions that disagree on a field
are raised to the user, resolution logged as a new entry.
- Rename .decision-log.md -> .memlog.md across SKILL.md and assets.
* core: add shared canonical memlog.py in src/scripts
Single source-of-truth memlog: append-only, chronological working-memory
log for skills. Installs to _bmad/scripts/memlog.py via the existing
src/scripts sync (beside resolve_customization.py), so any skill can call
it at runtime — bmad-spec is the first consumer.
Merges the neutral API (--workspace, free-form --type/--by, generic set)
with crash-safe fsync atomic writes. No lifecycle status by design: a
memory log records completion as an event entry, never a frontmatter flag.
Also accepts --path for callers that hold the file path directly. 30 tests.
* bmad-spec: include event in memlog --type list
The documented append --type set omitted event while the next line
requires --type event for terminal moments. Align the list.
* Fix memlog Python floor and exclude tests from install
- memlog.py: add 'from __future__ import annotations' so PEP 585/604
hints stay lazy; the script runs on Python 3.8+ instead of crashing
below 3.10. Correct the requires-python header to >=3.8.
- installer.js: filter tests/, __pycache__/, .pytest_cache/, and *.pyc
out of _installSharedScripts so dev-only files never ship to users.
The "rendered N files" progress line was pure diagnostic noise. The shim
already tells the LLM to ignore stderr and follow the stdout instruction, so
on success render.py now prints only the "read and follow ..." line.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The shim called bare `python`, which can resolve to Python 2 or be absent;
render.py needs 3.11+ for tomllib. Spell out python3 and the version
requirement. Also make the exit code authoritative: on a non-zero exit
(including an uncaught crash that writes only to stderr), do not proceed --
report what was printed and stop.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
render.py now merges the three customize layers (customize.toml ->
custom/bmad-quick-dev.toml -> .user.toml) with the same structural rules as
resolve_customization.py and inlines the resolved [workflow] values, so no
{workflow.*} placeholder survives. workflow.md drops its Step 1 runtime
resolver + manual-merge fallback; step-05 and step-oneshot drop their runtime
workflow.on_complete calls. The shared resolve_customization.py and every
other skill are untouched. Smoke test extended with a [workflow] override
fixture covering inlining, array append, and no-leak assertions.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The bare `python render.py` shim assumes the agent's working directory is
the skill directory, but agents run from the project root, so the script
is not found. Reference it as `{skill-root}/render.py` — BMAD's standard
token for a skill's installed directory, already used by every other
skill's resolve_customization.py invocation — and add the one-line
`{skill-root}` explainer so the model resolves it from an instruction
rather than guessing. Interpreter stays `python`; the python vs python3
choice is a separate cross-platform concern.
Load the four config layers through a load_toml helper that marks the
base _bmad/config.toml as required. A missing, unparseable, or unreadable
base now prints a HALT directive to stdout and exits, instead of being
silently skipped and then crashing downstream with a KeyError when a
derived value (e.g. implementation_artifacts) is absent. Optional layers
still warn on stderr and fall back to empty. Merge semantics are
unchanged (dict-aware deep merge, override wins for lists and scalars).
New test/test-quick-dev-renderer.js spins up a temp project with
base _bmad/config.toml and a _bmad/custom/config.user.toml override,
runs render.py, and asserts the override wins in rendered workflow.md
and that sprint_status is rooted at an absolute path in the temp
project. Registered as test:renderer in package.json and chained
into the npm test script.
Part of plan-quick-dev-python-config-hardening.md (F7).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The previous INSTALL_ONLY_PATHS entry 'render/' was a blanket prefix
that let every {project-root}/_bmad/render/... reference in any skill
slip past validation. Narrow to 'render/bmad-quick-dev/' so only this
skill's render buffer is whitelisted. Future skills adopting the
stdout-dispatch renderer pattern add their own entries explicitly.
Part of plan-quick-dev-python-config-hardening.md (F6).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
render.py rebuilds from scratch per the docstring, but
makedirs(exist_ok=True) only overwrites files that still exist in
the source — stale outputs from renamed/deleted source files linger
in _bmad/render/bmad-quick-dev/ forever. Remove every .md in the
render dir before the render loop; keep the dir itself and any
non-.md files.
Part of plan-quick-dev-python-config-hardening.md (F5).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Python text-mode open() with the platform default performs universal-
newline translation: on Windows, LF source files get written as CRLF,
producing spurious diffs when rendered output is compared against
source. Pass newline="" on both the source read and the rendered
write so line endings pass through verbatim.
Part of plan-quick-dev-python-config-hardening.md (F4).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
On Windows, os.path.join returns backslash-separated paths that can
misrender as escape sequences when later concatenated into POSIX
shell strings or regexes. Normalize the project root to forward
slashes after find_project_root, and use posixpath.join for every
path that gets baked into rendered .md files or joined into config
values. os.makedirs and os.listdir accept forward-slash paths on
Windows, so their call sites stay as-is.
Part of plan-quick-dev-python-config-hardening.md (F3).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Single happy path: central _bmad/config.toml with four-layer merge,
Python 3.11+ required (no ImportError guard), HALT if config missing.
Deletes load_flat_yaml, the YAML fallback branch, the setdefault block
for planning_artifacts/implementation_artifacts/communication_language,
and the tomllib ImportError fallback.
Part of plan-quick-dev-python-config-hardening.md (F0).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Move compile-time variable substitution out of the LLM and into a
deterministic Python step. SKILL.md becomes a two-line stdout-dispatch
shim that runs render.py and follows the instruction it prints. The
renderer reads BMad configuration from the central four-layer TOML
surface introduced in #2285 (_bmad/config.toml plus config.user.toml
and the two _bmad/custom/ overrides), with a fallback to the legacy
per-module _bmad/bmm/config.yaml for pre-#2285 installs.
Compile-time refs ({{.var}}) get substituted at render time. LLM-runtime
refs ({var}) pass through untouched.
Renderer (render.py)
- Python 3 stdlib only (tomllib, already bundled since 3.11). UTF-8 I/O.
Every invocation rebuilds from scratch — no hash, no cache.
- find_project_root walks up from cwd; HALT to stdout if no _bmad/
is found anywhere on the path.
- load_central_config deep-merges the four TOML layers in priority
order (base-team → base-user → custom-team → custom-user) so user
overrides in _bmad/custom/config.user.toml win over installer-
regenerated base values. flatten_central_config lifts scalar keys
from [core] and [modules.bmm] into the renderer's flat namespace;
module keys beat core on collision (matches the installer's own
core-key-stripping behavior).
- When _bmad/config.toml is absent, falls through to the legacy
flat-YAML parser for _bmad/bmm/config.yaml — the renderer keeps
working across the #2285 transition.
- {{.var}} substitution; unresolved refs emit empty string (Go
missingkey=zero semantics).
- Smart defaults for planning_artifacts / implementation_artifacts /
communication_language applied after config load. Derives
sprint_status / deferred_work_file from implementation_artifacts.
{{.main_config}} points at whichever surface was actually read.
- Renders every .md in the skill dir except SKILL.md to
{project-root}/_bmad/render/bmad-quick-dev/.
- On success, stderr summary plus a single stdout line:
"read and follow {workflow_md}". On failure, stdout HALT directive —
per the Anthropic skills spec, script stdout is the defined agent-
communication channel.
Skill entry (SKILL.md)
- Two-line shim: run python render.py, follow stdout. No template
tokens in SKILL.md itself.
Template conversions
- workflow.md, step-01..05, step-oneshot, sync-sprint-status: convert
every compile-time {var} reference to {{.var}}. Runtime refs
preserved.
- spec-template.md untouched (single-curly comment hint stays as
documentation).
Skill-prose cleanups bundled in
- Remove dead step-file frontmatter: empty-string variable declarations
(spec_file, story_key, diff_output, review_mode) in quick-dev step-01
and code-review step-01; empty --- --- blocks in step-03 and step-05;
the specLoopIteration counter init moved from step-04 frontmatter into
the step body where first-entry vs loopback semantics are explicit.
- Unify the language rule across all six quick-dev step files plus
workflow.md.
Tooling
- tools/validate-skills.js: add TPL-01 rule. Files whose name contains
"template" must not contain compile-time {{.var}} substitutions.
Template files seed durable, version-controlled artifacts that
execute on other machines; baking a value at render time would
freeze a machine-local path into every downstream artifact.
- tools/validate-file-refs.js: add render/ to INSTALL_ONLY_PATHS so
the validator recognizes the runtime-generated buffer.
- tools/skill-validator.md: document TPL-01; deterministic rule count
bumped from 14 to 15.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The Blind Hunter subagent intentionally has no tool access, but the
review steps never said how {diff_output} should be delivered. Orchestrators
typically wrote the diff to a temp file and asked the agent to read it,
which silently fails (0 tool calls), and the 10-finding requirement then
pushes the agent to hallucinate findings against code it never saw.
State explicitly that the diff is passed inline in the subagent prompt,
in both bmad-code-review step 2 and bmad-quick-dev step 4.
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
* docs(fr): translation of install-custom-modules
Reference commit 97d32405
* docs(fr): refinement of forensic-investigation
* docs(fr): translation of customize-bmad TOML customization rewrite
Reference commits 0dbfae67, 4405b817, ffdd9bc6, b63086f2"
* fix(docs): handle non-ASCII anchors in link validator
Anchor validation failed for links containing accented characters
(e.g. ./customize-bmad.md#dépannage) because the raw anchor didn't
match the slugified version produced by extractAnchors.
Normalize anchors through decodeURIComponent + headingToAnchor before
comparing, and guard against malformed URI components.
* docs(fr): translation of install-bmad channel and config rewrite
Reference commits 3d824d4c, 91a57499, 0f852a38
* docs(fr): translation of expand-bmad-for-your-org organizational customization patterns
Reference commits c52c9b5b, b63086f2, 4405b817, 0dbfae67
* docs(fr): update install-custom-modules
Reference commit 231a2036
* docs(fr): consolidate non-interactive installation into unified install-bmad guide
Replace standalone non-interactive-installation.md with a redirect stub
pointing to the Installations CI non interactives section in install-bmad.md.
* docs(fr): translation of named-agents
Reference commits 0dbfae67, 4405b817, b63086f2
* docs(fr): refinement of upgrade-to-v6
* docs(fr): refine agents.md
* docs(fr): refine commands.md
rename bmad-create-prd to bmad-prd and update skill descriptions
* docs(fr): refine workflow-map-diagram
Reference commit c52c9b5b
rename create-prd to prd, create-product-brief to product-brief
add prfaq workflow, update agent labels and output names
refine French wording throughout
* docs(fr): update and refine workflow-map
Reference commits: 380590ac52c9b5
* docs(fr): update and refine getting-started
Reference commits c52c9b5b, 0f852a38
rename bmad-create-prd to bmad-prd, add PRD intents section
update Quick Reference table, refine French wording throughout
* docs(fr): refine index.md
Reference commit 0dbfae67
refine French wording throughout, improve phrasing and table formatting
* docs(fr): apply French typographic conventions across all docs
regex-based pass followed by AI + manual review of all 34 source files
Rules applied:
- Apostrophe: ASCII ' → curly ’ (U+2019) in all French prose
- Guillemets: ASCII "..." → « … » with narrow no-break space (U+202F) on both sides
- Narrow no-break space (U+202F): before ; ? ! and after « / before »
- No-break space (U+00A0): before : in French prose
- Thousands separator: narrow no-break space (U+202F) in 4+ digit numbers
Additional review fixes: remaining ASCII quotes in _STYLE_GUIDE.md
checklist items, testing.md, and party-mode.md numbering.
Preserved exclusions: YAML frontmatter delimiters, code blocks,
backtick inline code, URLs, footnote syntax, and English UI text.
* docs(fr): align sidebar ordering with current English docs
Update sidebar order values across all French explanation and how-to
pages to match the live English documentation structure.
* docs(fr): fix omission in quick-dev from english
* docs(fr): style guide formatting
* docs(fr): use quick-dev wording in workflow-map-diagram-fr
* docs(fr): fix typos
* docs(fr): add bmad-investigate / IN trigger to agent tables
The forensic investigation feature added the IN menu trigger and
bmad-investigate skill, but the French docs that enumerate triggers
and agent capabilities were not updated.
- agents.md: add IN trigger and Enquête de code to Amelia's row
- named-agents.md: add Enquête de code to Amelia's capabilities
* docs(fr): fix agent skill identifiers to use bmad-agent-* prefix
The agent skill identifiers in agents.md and commands.md were missing
the -agent- segment of the namespace (e.g. bmad-pm instead of
bmad-agent-pm). All agent launchers use the bmad-agent-* naming
convention since the installer generates skill directories under that
prefix.
- agents.md: fix bmad-dev, bmad-analyst, bmad-pm, bmad-architect,
bmad-ux-designer, bmad-tech-writer
- commands.md: fix bmad-pm, bmad-architect
* docs(fr): rename bmad-create-ux-design to bmad-ux (#2413)
Apply ee47e30c (refactor(bmad-ux): spine-based UX skill) to French docs.
Rename skill bmad-create-ux-design → bmad-ux and update outputs
from ux-spec.md to DESIGN.md + EXPERIENCE.md.
* docs(fr): translate bmad-spec section
French translation of the bmad-spec section introduced in aa6dece
(feat(bmad-spec): introduce Spec kernel distiller skill (#2417)).
* docs(fr): improve core-tools locution, phrasing and typography
Broader pass across all sections of core-tools.md for more idiomatic
French: consistent section headers (À utiliser quand, Fonctionnement),
natural verb choices, fluid sentence construction and corrected
punctuation.
* docs(fr): apply French typography and table formatting pass
Continuation of 27002100. Systematic pass across all French documentation
assisted by an automated French typography linter:
- Replace regular space with NBSP (U+00A0) before colons per French
typographic convention
- Align table separator rows to match column widths
- Fix thousands separator in install-bmad.md (5000 → 5 000)
- Correct glossary example code block rendering in _STYLE_GUIDE.md
* docs(fr): fix missing french typography on roadmap.mdx
* docs(fr): translate web-bundles explanation and how-to
French translation of:
- docs/explanation/web-bundles.md
- docs/how-to/use-web-bundles.md
Reference commits: 7729ad46, d659a03d, 3bc2ad30
* docs(fr): refresh skill metadata references
Fixes#2437 for French.
- agents.md: update PM triggers CP/VP/EP → PRD, remove stale US trigger
from Technical Writer, align PRD description to create/update/validate
- commands.md: fix Cursor/Windsurf skill paths to .agents/skills/,
update core tools count to 12, align PRD description
- core-tools.md: add missing bmad-customize tool entry and section with
link to customize-bmad how-to
- party-mode.md: replace stale "BMad Master orchestre" with "Le Party
Mode orchestre la discussion"
---------
Co-authored-by: Brian <bmadcode@gmail.com>
* Initial draft: brainstorming improved
* bmad-brainstorming: apply quality-analysis fixes
- Add ## Overview heading (prose was already present)
- Fix resume scan to glob output_dir/*/session.md (per-session subfolders);
handle multiple in-progress sessions
- Anchor brain_methods to {skill-root} and always pass --file to brain.py
(--file before subcommand); drop the unreliable conditional
- Log decisions, dismissed E3 false positive, verified tests (15 passed)
* bmad-brainstorming: memlog session memory + facilitation refinements
Replace the running-log concept with a generic, append-only memlog
(scripts/memlog.py): a flat, chronological, write-only session memory any
skill can reuse. Entries land at the end in the order they happen; --type
tags the kind (idea/insight/question/decision/technique); nothing is
grouped, reordered, or rewritten. The file is .memlog.md, read only on resume.
- scripts/memlog.py (init/append/set) + test_memlog.py (20 tests)
- SKILL.md: reordered into framing + flow; lean Memlog framing; batch
technique model (Facilitator Chosen / Browse / Category / Inventive Flow);
Progressive removed; facilitation stance preserved
- references/finalize.md: two-move synthesis + opt-in artifacts, each derived
from the memlog via subagent
- references/headless.md, customize.toml: memlog wiring + per-topic folders
* bmad-brainstorming: gate technique-flow choice and stop full-catalog dumps
- brain.py: `list` now requires --category or --all; a bare `list` is refused so the full ~100-technique catalog can no longer flood context. --all is the deliberate full-dump escape hatch.
- SKILL.md: technique-flow selection is now a hard gate (present the four ways, wait for the user's pick) instead of a soft default that got skipped; Stance "no multiple-choice" rule scoped to generation, with an explicit carve-out for that one process menu.
- headless.md: use `list --all` deliberately, passing --file.
- customize.toml: fix stale "progressive flows" wording to match the four real flows.
- test_brain.py: regression tests for the list guard (bare refused, --all dumps all).
* bmad-brainstorming: remove .decision-log.md build-time artifact
The workflow-builder writes .decision-log.md to track its own build session; it is a build-time artifact, not part of the shipped skill, so it should not be committed to the project.
* bmad-brainstorming: facilitation modes, attributed memlog, and the selection composer
- Three facilitation modes chosen up front (Facilitator / Creative Partner / Ideate for me), each a loaded frame; SKILL.md routes mode + technique selection primarily through the selection page, with the in-chat menus as fallback.
- memlog: optional --by user|coach attribution (required in Creative Partner) so authorship stays visible in the log.
- brain.py: generates a self-contained "browse all" selection page (brain-selector.html) - a session composer with facilitation mode, a hand-picked + Random + Invent + AI-picks technique strategy, category toggle-chips and a category-aware filter, and a copy-to-clipboard prompt with a paste-back banner. Category-tinted cards, 13 crafted category icons, and a hand-assigned icon for each of the 100 techniques. `html` writes to a file (never dumps the catalog into context); a snapshot test keeps the shipped page in sync with the CSV.
- Drop the now-unused Six Thinking Hats detail file; the catalog needs no detail files.
- finalize.md: synthesis is mode-aware and the "hand them the mirror" step reads the by-attribution tags.
* bmad-brainstorming: fix code-review findings
memlog.py
- Parse frontmatter by the first line that is exactly `---`, so a `---` inside a
topic/goal value no longer truncates the block, drops `status`, and breaks resume
forever. Neutralize newlines in field values on render too.
brain.py (selector page + CLI)
- Composer: category toggles now define session scope; the text filter is a pure
browse aid. checked() and the random pool both key off scope (offCats), so hidden
cards are never silently copied and a stray filter term can't starve a random draw.
- Clipboard: only show the "Copied!" banner when the copy actually succeeds; on
failure show a warning and a prefilled prompt() so the text is never lost.
- category_style: fall back to the neutral glyph instead of KeyError if the hue/glyph
dicts ever desync.
- random: clamp -n so a negative/oversized value returns cleanly instead of crashing.
- --extra: merge a JSON overlay of additional_techniques into every command, so the
browse page and category draws include custom techniques/categories as advertised.
docs
- SKILL.md: fix dangling `## Choosing Your Mode` anchor and the "Copy selection"
button label; document --extra in the regen instructions.
- mode-autonomous.md: persist the mode flip when handing off from autonomous, so a
resume restores the new stance.
- finalize.md: grammar/typo fixes (CodeRabbit).
tests
- Regression tests for the memlog `---` fix, --extra merge, negative -n, and the
category fallback; regenerated the snapshot-tested selection page. Renamed the
shadowing `type`/`l` locals flagged by CodeRabbit. 52 passing.
* bmad-brainstorming: composer header polish + dark mode
- Center header content (.hwrap) so it aligns with the card column on wide screens.
- Replace the text filter with jump-nav: category chips smooth-scroll to their
section (offset for the sticky header); drop the category exclude-toggle, so
Random/AI draw from the whole catalog.
- Fix narrow-screen crowding between the chips and the Copy prompt button.
- Move Copy prompt to the end of the Techniques row, anchored to the Total readout.
- Add a per-mode hint line that explains the selected facilitation stance.
- Dark mode: refactor all colors to CSS variables + a dark palette, with a header
toggle (☾/☀) that defaults to system preference and persists in localStorage; an
inline head script applies the theme before first paint to avoid a flash. Category
hues are lifted toward white on dark surfaces to stay legible.
Regenerated the snapshot-tested selection page; SKILL.md wording updated (chips are
jump-nav, not a filter). 52 Python tests passing.
* bmad-brainstorming: condense SKILL.md, extract resume + in-chat technique frames
- Rewrite SKILL.md: consolidate framing, merge session setup into Run a
Session, descriptive (non-imperative) Overview true across all 3 stances
(~2,790 -> ~2,070 tokens)
- Extract Resuming to references/resume.md (loads only on resume)
- Extract in-chat technique selection to references/in-chat-techniques.md
(loads only when the composer page is declined)
- Add HTML-open recovery guidance to the composer-page step
- Ideate-for-me now auto-produces the HTML keepsake (finalize.md +
mode-autonomous.md) instead of asking first
* Enhance bmad-brainstorming: goal facet, proven grouping, convergence, icon sidecar
Catalog (brain-methods.csv -> 108 methods):
- Add provenance / good_for / audience columns (additive, backward-compatible)
- Add 8 researched classic methods: How Might We, Job to Be Done, Empathy Map,
Backcasting, TRIZ Contradiction, Fishbone Diagram, Build on What Works, Scenario Cross
Selector page (brain.py generator):
- "Proven & Professional" lead group (29 named methods, cross-category)
- Super-group ordering (Structured/Creative/Wild/Introspective) replacing alphabetical
- "Great for" goal filter driven by good_for tags
- Per-category "Invent a ... technique" cards reusing the invent flow
Convergence:
- New references/converge.md (diverge -> converge -> finalize); wired into SKILL.md
Maintainability:
- Extract category + technique icons to assets/brain-icons.json; brain.py loads the
sidecar, with logic and fallbacks staying in code (8 new icons added, full coverage)
Docs:
- Add analysis/ (catalog-analysis.md + method-matrix.csv): the 4-axis review behind these changes
All 52 tests pass.
Reframes the party-mode skill around outcomes instead of a rigid
subagent script, addressing issue #2280.
- Voicing the room is the default; subagents become opt-in (--subagents)
for rounds that genuinely need independent thinking.
- Removes the lossy 400-word context cap and the 'Do NOT use tools'
constraint that drove subagents to fabricate on grounded questions.
- Adds a 'What Good Feels Like' bar: short in-character turns, real
drama over consensus, brevity by default.
- Adds orchestrator weaving so independently-produced turns read as one
conversation without altering what any persona actually argued.
- Slims SKILL.md from 128 to 75 lines.
With shallow clones (--depth 1), `origin/HEAD` becomes stale after the
initial clone. The update path used `git reset --hard origin/HEAD` which
never picked up new commits pushed to the default branch.
Resolve the default branch name via `git symbolic-ref refs/remotes/origin/HEAD`,
then fetch and reset against `origin/<branch>` explicitly. Falls back to
`main` if origin/HEAD is not set.
Co-authored-by: Brian <bmadcode@gmail.com>
loadExistingConfig only read from legacy _bmad/<module>/config.yaml files, but
the installer writes user-scoped answers (user_name, communication_language, etc.)
to _bmad/config.user.toml. On every subsequent reinstall those values were not
loaded back, so the user got re-prompted instead of seeing their prior answers as
defaults.
Adds parseCentralToml — a lightweight line scanner matching the installer's own
TOML output format — and updates loadExistingConfig to read config.toml and
config.user.toml first (merging both into the same section buckets). Legacy
per-module config.yaml files are kept as a fallback for pre-v6 installations.
Co-authored-by: RobertOcsko <robert.ocsko@;seon.io>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: Brian <bmadcode@gmail.com>
* fix(installer): refresh custom-source cache on quick-update and persist channel marker
* fix(installer): persist real next ref and atomically dedupe custom refresh
* fix(installer): preserve custom-source cache when remote unreachable
When git fetch fails against an existing custom-module cache, cloneRepo
previously wiped the cache and attempted a fresh clone, which then also
failed for the same reason (network down, repo deleted/moved, auth
revoked) — leaving the user with no usable cache. With the new
quick-update refresh path calling cloneRepo for every cached custom
module, this turned transient remote outages into cache loss on every
quick-update.
- cloneRepo: on fetch failure with an existing cache, keep the previous
clone and surface a warning via prompts.log.warn instead of removing
the cache. The downstream metadata write uses the existing HEAD.
- _refreshRepoCacheOnce: update the comment to reflect that the common
"remote unreachable but cache exists" case is now handled inside
cloneRepo; warn on the remaining unrecoverable failures so they
aren't silent.
Tests: 349 passed, 0 failed.
---------
Co-authored-by: Brian Madison <bmadcode@gmail.com>
Make bmadcode.com/web-bundles/ the single supported install path. The
site keeps install steps current as Gemini and ChatGPT evolve, always
points at the newest tagged release, and turns one signup into the
notification channel for new bundles.
- README.md: drop direct-folder install reference, point to the site
- web-bundles/README.md: lead with Install (site), reframe folder as
source for the shelf rather than install target
- docs/explanation/web-bundles.md: replace the per-bundle INSTRUCTIONS.md
steer with the site, replace the dead how-to link
- docs/how-to/use-web-bundles.md: rewrite as a short pointer to the site
(kept the file so existing inbound links resolve), retain prerequisites,
persona customization, and build-your-own sections
* feat(web-bundles): add release packager + bundle manifest
Adds the infrastructure for shipping web bundles as downloadable ZIPs
attached to a GitHub Release, consumed by the upcoming
bmadcode.com/web-bundles/ page.
- web-bundles/bundles.json — manifest with persona, tagline, description,
accent color, motif key, knowledge files, and feature flags
(web-browsing, deep-research, stitch integration) for each of the 6
bundles. Top-level releaseTag and downloadUrlPattern so the
consuming page can construct download URLs without hardcoding.
- tools/bundle-web-bundles.js — packager that zips each bundle dir into
dist/web-bundles/{slug}.zip and prints the gh release create command.
Zero dependencies; uses system zip.
- .gitignore — exclude dist/web-bundles/ build artifacts.
The web-bundles-v1.0.0 release on GitHub is currently in draft state
with the 6 zips attached; it'll be published in coordination with the
Ghost site page going live.
* fix(web-bundles): single-source release tag, sharper bundle copy
- Remove downloadUrlPattern from bundles.json — the consuming page
derives the URL from releaseTag, so version bumps now touch one
field instead of two.
- product-brief-coach: drop "one-page" (briefs are whatever length
the product earns).
- brainstorming-coach: real numbers — 60 techniques across 10
categories — with concrete examples (SCAMPER, Drunk History
Retelling, Nature's Solutions, Six Thinking Hats, etc.) so the
card actually communicates the surprising breadth.
* fix(web-bundles): harden release script per PR review
- Verify the zip CLI is on PATH up front with a clear install
hint, instead of crashing mid-zip with an opaque execSync error.
- Wrap JSON.parse in try/catch; validate the manifest shape (bundles
array non-empty, releaseTag present, slug present per entry) before
trying to package, so config errors fail with a targeted message.
- Catch zip failures per-bundle and surface the failing slug.
- Refuse to print the gh release command when zero bundles were
packaged (would otherwise mislead the user into creating an empty
release).
- Derive --title from manifest.releaseTag so the printed command can
never drift from the actual tag (was previously hardcoded
"Web Bundles v1" while the tag had moved to v1.0.0).
- Remove the stale `web-bundles-v1` example from the file header.
Addresses augmentcode bot review comments on PR #2424.
* docs(web-bundles): rewrite copy to actually sell what each bundle does
The JSON drives the bmadcode.com/web-bundles/ page; previous copy
was generic and undersold the actual capabilities. Rewrote each
tagline + description to lead with concrete, differentiating facts
pulled directly from each bundle's SKILL.md:
- Brainstorming Coach: 60 techniques across 10 categories with
specific names (SCAMPER, Drunk History Retelling, Nature's
Solutions, Shadow Work Mining, Superposition Collapse); calls
out the 4 routes (browse, recommend, random, progressive) and
the ~100-idea quantity-unlocks-quality target.
- Product Brief Coach: names the three intent modes (Create /
Update / Validate) and the two working paths (Fast / Coaching);
surfaces the [ASSUMPTION] tag system and the Addendum.
- PRFAQ Coach: details the 4 stages (Ignition / Press Release /
Customer FAQ / Internal FAQ + Verdict), the 9 press release
sections, the weasel-word list ("best-in-class", "seamless"),
and that it adapts for commercial, internal, OSS, community.
- PRD Coach: spells out the two entry points (Vision+Features
vs Journey-led), named-protagonist journeys, glossary
discipline, stable ID system (FR-1..N, SM-C1..N), and the
7-dimension validation rubric.
- UX Coach: leads with the two-spine contract (DESIGN.md +
EXPERIENCE.md), Don Norman framing, named-protagonist
journeys, surface closure as the test, and Stitch integration.
- Market & Industry Research: leads with Deep Research as the
engine, names Porter and Christensen as anchors, lists the 6
deliverable sections, and frames the deliverable as synthesis
not a research dump.
* fix(web-bundles): security hardening + strict bundle validation
Two issues raised by coderabbit on the latest commit:
1. Shell injection surface: execSync was building the zip command
with a template literal that interpolated bundle.slug from JSON.
Even with our controlled inputs, a slug with shell metacharacters
would break quoting. Switched to execFileSync with an argument
array (no shell) and added a strict ^[a-z0-9][a-z0-9-]*$ slug
regex enforced before any FS or zip call.
2. Missing bundle directories were [SKIP]-warned but the script
still printed the release command, allowing an incomplete release
to ship cleanly. Now treated as fatal: any missing or invalid slug
blocks the printed gh command and exits non-zero with the offending
slugs listed.
* feat(docs): add sidebar order validator
Adds tools/validate-sidebar-order.js to validate sidebar.order values
in YAML frontmatter across English and translated docs.
Checks for duplicate orders, gaps in sequence, and missing order fields.
For translations, also warns on order drift from English counterparts.
Wired into the quality script as docs:validate-sidebar.
* fix(validate-sidebar): tighten language detection and drift guard, add docstrings
* fix(validate-sidebar): replace subdirectory heuristic with locale pattern matching
detectLanguageDirs() previously classified any top-level docs/ directory
containing subdirectories as a translation language. This was too broad —
if an English section ever gained nested subfolders it would be silently
excluded from validation.
Replaced with a BCP 47 locale-code regex (/^[a-z]{2}(?:-[a-zA-Z]{2})?$/)
that matches known patterns (cs, fr, vi-vn, zh-cn) and won't falsely
classify content sections like explanation/ or reference/.
* fix(validate-sidebar): guard drift check against undefined order values
extractSidebarOrder() returns { hasSidebar: false } when no sidebar block
exists, leaving order as undefined rather than null. The drift check only
guarded against null, allowing undefined values to emit noisy warnings
like "Order drift: ... order undefined".
Changed the guard to typeof === 'number' which correctly excludes both
undefined and null without relying on a specific sentinel value.
* chore(validate-sidebar): add JSDoc docstrings to all functions
Adds @param and @returns annotations to extractSidebarOrder,
detectLanguageDirs, getEnglishSections, checkDirectory,
checkTranslationDrift, and relativePath.
* fix(validate-sidebar): add to pre-commit hook
* refactor(validate-sidebar): harden parsing and edge-case handling
Refactor to main() wrapper with pure return-based APIs, single directory
scan, and shared reporting. Harden frontmatter parsing (anchored delimiter,
direct-child-only order extraction, flow mapping support) and validation
(Infinity/zero guard, gap flood cap, multi-segment locales, graceful ENOENT).
* docs: fix sidebar.order duplicates and gaps across all locales
Resolves all validator errors flagged by the new
tools/validate-sidebar-order.js check.
English (docs/{explanation,how-to,reference}/):
- Renumbered to remove duplicates; established reading order
for new explanation pages added since orders were last set.
Translations (cs, fr, vi-vn, zh-cn):
- Mirrored English structural ordering where files exist, then
compacted to 1..N within each directory to eliminate gaps
caused by missing translation files.
Non-blocking drift warnings remain where translation directories
have fewer files than English; these are expected per the
validator's design.
---------
Co-authored-by: Brian Madison <bmadcode@gmail.com>
* fix(skills): strengthen activation guardrails for all workflow skills
Add explicit "Activation is complete" boundary markers that require
confirming activation_steps_prepend and activation_steps_append were
fully executed before beginning the main workflow.
Previously, the guardrail was either missing (bmad-product-brief,
bmad-prd, bmad-investigate) or too weak ("Begin the workflow below").
LLM agents would short-circuit complex activation sequences (INCLUDE →
READ → RUN → CHECK → FILTER → CD) by guessing variables instead of
executing steps in order, causing append steps and on_complete hooks
to be silently skipped.
The new guardrail explicitly names both prepend and append steps,
requiring confirmation before proceeding. This prevents agents from
starting the main workflow in parallel with activation.
23 skills updated: bmad-product-brief, bmad-prd, bmad-prfaq,
bmad-investigate, bmad-create-story, bmad-dev-story,
bmad-quick-dev, bmad-code-review, bmad-correct-course,
bmad-sprint-planning, bmad-sprint-status, bmad-retrospective,
bmad-qa-generate-e2e-tests, bmad-checkpoint-preview,
bmad-check-implementation-readiness, bmad-create-architecture,
bmad-create-epics-and-stories, bmad-generate-project-context,
bmad-create-ux-design, bmad-document-project, bmad-market-research,
bmad-technical-research, bmad-domain-research.
* fix(skills): extend activation gate to agent + new skills, refine placement
- bmad-product-brief / bmad-prd: pull activation_steps_append out of
the numbered list so the sentinel reads as a paragraph break, not
as the next list item.
- bmad-investigate: move the sentinel above Step 7 (routing) — Step 7
is workflow routing, not activation; the gate must fire first.
- bmad-agent-{analyst,tech-writer,pm,ux-designer,architect,dev}: add
the same gate between Step 7 (append) and Step 8 (menu dispatch).
Persona skills had the same short-circuit risk but no sentinel.
- bmad-ux, bmad-spec: new skills introduced on main after this branch
forked; apply the same gate so the pattern stays consistent.
- removals.txt: register bmad-create-ux-design as renamed to bmad-ux.
---------
Co-authored-by: Brian Madison <bmadcode@gmail.com>
- Drop "one-click install" framing in README and explanation page; setup is
multi-step but consistent across the shelf.
- Drop "two files (sometimes three)" claim; honestly name SKILL.md,
INSTRUCTIONS.md, and any required data files (templates, CSVs,
validation checklists).
- Add explicit setup pattern in README (create Gem/GPT, upload knowledge,
paste instructions, save).
- Add "Updating and customizing" section to the explanation page covering
re-upload-the-attachments updates and the rule of thumb that custom
changes belong in the pasted instructions block, not the knowledge
files, so future updates don't clobber team customizations.
* feat(web-bundles): bring back V4 web bundles for V6
Restores the V4 web bundle pattern for V6. BMad planning skills are
repackaged as one-click installs for Google Gemini Gems and ChatGPT
Custom GPTs, so users can run analysis and planning conversations in
flat-rate web LLM subscriptions instead of metered IDE tokens.
Current shelf (6 bundles): brainstorming, product brief, PRFAQ, PRD,
UX, market and industry research. Each bundle ships a SKILL.md
protocol, an INSTRUCTIONS.md with a default persona and a contrasting
swap example, and any required data files.
New in this commit:
- Market & Industry Research bundle merging market and domain research
with Porter and Christensen anchors, Mary persona inherited from the
BMad analyst agent, Geoffrey Moore swap example, Deep Research mode
integrated as the default research path
- web-bundles/README.md folder index listing all 6 bundles
- README.md section announcing the V6 web bundle shelf
- docs/explanation/web-bundles.md concept page (what, why, when)
- docs/how-to/use-web-bundles.md install steps for Gemini Gems and
ChatGPT Custom GPTs
* docs(web-bundles): unindent admonition closer in use-web-bundles.md
The :::note[Prerequisites] closer was indented under the last bullet,
which can prevent the admonition from closing and break Starlight
rendering for the rest of the page. Flush left now.
The forensic investigation feature (commit 380590aa) added the IN menu
trigger and bmad-investigate skill, but several docs that enumerate
triggers and agent capabilities were not updated.
- agents.md: add IN trigger and Forensic Investigation to Amelia's row
- named-agents.md: add forensic investigation to Amelia's capabilities
Co-authored-by: Brian <bmadcode@gmail.com>
Alex Verkhovsky
Loic Duong
Dov Benyomin Sohacheski
SLUR
Davor Racic
Brian
PinkyD
Aristo Rinjuang
Oneby Wang
Emmanuel Atsé
duliangang
github-actions[bot]
hanhnt2-hblab
Jacob du Toit
Jérôme Revillard
robertocsko-seon
SevenSteven
Farzad Rashidi