bd1c0053d5
Three-layer customization (skill defaults → team → user) for BMad agents
and any skill that opts in. Users edit `_bmad/custom/{skill-name}.yaml`
(team, committed) or `{skill-name}.user.yaml` (personal, gitignored);
customizations survive updates.
Resolver is a Python script using PEP 723 inline metadata, invoked via
`uv run` so deps auto-install into a cached isolated env on first call.
This aligns with Anthropic's Agent Skills spec and BMB conventions, and
keeps the dependency declared (scannable by pip-audit/Dependabot) rather
than vendored.
## Design choices
- **Agent identity is hardcoded** in SKILL.md (name, title, Overview prose)
so skills can be invoked reliably by role *or* default name. Brand
recognition is preserved; customization shapes behavior, not identity.
- **Luminary-anchored personas** (e.g. "Channels Martin Fowler's
pragmatism and Werner Vogels's cloud-scale realism") deliver ~55%
token savings per agent while preserving distinctive voice beats.
- **Universal per-field merge rules** with v6.1-compatible agent
semantics: metadata shallow-merge, persona replace, critical_actions
and memories append, menu merge-by-code, all else deep-merge.
- **Workflow customization** shares the same surface — `bmad-product-brief`
pilots `activation_steps_prepend`, `activation_steps_append`, and
`skill_end` hooks that any workflow-style skill can adopt.
## Infrastructure
- `_bmad/scripts/` houses shared Python scripts (resolver + future).
- `_bmad/custom/` is provisioned empty with a seeded `.gitignore` for
`*.user.yaml` on fresh installs.
- Installer filters ensure `scripts/`, `custom/`, and sidecar-generated
`memory/` directories are never treated as modules.
- Dead v6.1 code cleaned up: `_config/agents/` no longer created,
`metadata.capabilities` removed from schema and CSV manifest.
|
||
|---|---|---|
| .. | ||
| commands | ||
| core | ||
| ide | ||
| modules | ||
| README.md | ||
| bmad-cli.js | ||
| cli-utils.js | ||
| file-ops.js | ||
| fs-native.js | ||
| install-messages.yaml | ||
| message-loader.js | ||
| project-root.js | ||
| prompts.js | ||
| ui.js | ||
| yaml-format.js | ||
README.md
BMad CLI Tool
Installing external repo BMad official modules
For external official modules to be discoverable during install, ensure an entry for the external repo is added to external-official-modules.yaml.
For community modules - this will be handled in a different way. This file is only for registration of modules under the bmad-code-org.
Post-Install Notes
Modules can display setup guidance to users after configuration is collected during npx bmad-method install. Notes are defined in the module's own module.yaml — no changes to the installer are needed.
Simple Format
Always displayed after the module is configured:
post-install-notes: |
Thank you for choosing the XYZ Cool Module
For Support about this Module call 555-1212
Conditional Format
Display different messages based on a config question's answer:
post-install-notes:
config_key_name:
value1: |
Instructions for value1...
value2: |
Instructions for value2...
Values without an entry (e.g., none) display nothing. Multiple config keys can each have their own conditional notes.
Example: TEA Module
The TEA module uses the conditional format keyed on tea_browser_automation:
post-install-notes:
tea_browser_automation:
cli: |
Playwright CLI Setup:
npm install -g @playwright/cli@latest
playwright-cli install --skills
mcp: |
Playwright MCP Setup (two servers):
1. playwright — npx @playwright/mcp@latest
2. playwright-test — npx playwright run-test-mcp-server
auto: |
Playwright CLI Setup:
...
Playwright MCP Setup (two servers):
...
When a user selects auto, they see both CLI and MCP instructions. When they select none, nothing is shown.