From hashi-blueprints
Use after modifying existing systems to update blueprint documentation. Read blueprints with read_blueprint before changes, update with write_blueprint after. Prevents documentation drift.
How this skill is triggered — by the user, by Claude, or both
Slash command
/hashi-blueprints:blueprints-maintenanceThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Documentation drifts from implementation when:
Documentation drifts from implementation when:
Read existing blueprint using MCP tools:
// Find the blueprint
const results = await search_blueprints({ keyword: "auth" });
// Read it to understand current documentation
const blueprint = await read_blueprint({ name: "authentication" });
Note what documentation exists:
Plan documentation updates alongside code changes
Re-read the blueprint to verify accuracy:
const current = await read_blueprint({ name: "authentication" });
Verify each section:
Update using MCP tool:
await write_blueprint({
name: "authentication",
summary: "Updated summary if needed",
content: updatedContent
});
Remove stale content - outdated docs mislead
When adding functionality:
When changing behavior:
When deprecating or removing:
When restructuring without behavior changes:
Signs blueprints need attention:
Prioritize by impact:
When reviewing blueprints:
## Verification Checklist
- [ ] Overview matches current purpose
- [ ] All public APIs documented
- [ ] API signatures accurate
- [ ] Examples execute correctly
- [ ] Behavior matches implementation
- [ ] File paths exist
- [ ] No removed features documented
- [ ] Related systems links work
- [ ] No duplicate content with other blueprints
/blueprints command to regenerateThe hashi-blueprints hooks automatically:
npx claudepluginhub thedotmack/han --plugin hashi-blueprintsGuides completion of development work by verifying tests, detecting environment, and presenting structured options for merge, PR, or cleanup.
Enforces test-driven development: write failing test first, then minimal code to pass. Use when implementing features or bugfixes.
Guides creation and editing of skills using test-driven development with pressure scenarios and subagents to verify agent compliance.