From spec-driven
Runs the change lifecycle when documented behavior changes: write an ADR for the why, edit the living spec to the new current state, bump governing_adrs, and supersede any reversed ADR.
How this skill is triggered — by the user, by Claude, or both
Slash command
/spec-driven:evolving-a-specThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
The end-to-end change flow that binds an ADR (the why) to a living-spec edit
The end-to-end change flow that binds an ADR (the why) to a living-spec edit (the new current state). Use it when both the decision and the documentation must move together.
authoring-living-spec.writing-adr.writing-adr to capture the why (proposed → accepted).authoring-living-spec to edit the affected spec(s) to the new current state.governing_adrs; bump last_updated.supersedes; old ADR → superseded + superseded_by).pnpm spec:ls / pnpm adr:ls to confirm the new state is listed.STOP: never edit spec behavior without a governing ADR.
| Change type | Sub-skills, in order |
|---|---|
| New behavior on an existing capability | writing-adr → authoring-living-spec (add ADR to governing_adrs) |
| Reversal of an accepted decision | writing-adr (new, supersedes) → flip old ADR → authoring-living-spec |
| Deprecation of a capability | writing-adr → authoring-living-spec (set spec status: deprecated, point to replacement) |
${CLAUDE_PLUGIN_ROOT}/references/change-lifecycle.mddocs/specs/meta/spec-governance.mdGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub baranovxyz/agent-context-kit --plugin spec-driven