repo-map-refresh
Skill20 starsChanged 3 months agoMITView on GitHub
Regenerate the auto-generated repo map the planner grounds its plans in. USER-ENTRY — invoke only when the user explicitly runs /repo-map-refresh; never autonomously, never from a subagent (guard-enforced).
In this repository7 other files it gives its agents.
- add-repoSkill
- dev-workflowSkill
- init-workspaceSkill
- migrate-workspaceSkill
- story-workflowSkill
- workflow-statusSkill
- workspace-configSkill
The file
Files
repo-map-refresh
The repo map is a navigation aid, never hand-maintained — corrections go
through regeneration (design.md piece 5B). /init-workspace and
/add-repo both point here for the identical generate-or-regenerate
procedure (step 2 below) — this is the one place it's maintained; don't
fork a second copy of it elsewhere.
${CLAUDE_PLUGIN_ROOT}/bin/harness repo-map-check --repo-name <n> --repo <path>— report missing / fresh / stale (+ commits behind) to the user.- To regenerate: spawn the planner shape with
harness-mode: repo-mapas the prompt's FIRST line (the spawn guard regex-matches this exact header — prose that merely mentions repo-map does not satisfy it; this is a declared out-of-run spawn,pipeline/surfaces.yaml'sout_of_run_spawns, legal regardless of whether other runs exist in the workspace) and the repo path. Preferrun_in_background: falsewhere the Agent tool has that parameter — the reply then flows back through PostToolUse. Where the tool defaults spawns to background or has no such parameter, the launch stub returns immediately and the platform's own completion notification is the ONLY signal: WAIT for it before step 3. Know that the harness cannot track THIS spawn in flight — capture early-returns for out-of-run spawns BEFORE the stub-recognition gate, so a backgrounded repo-map spawn records nospawn-pending, appears nowhere inshow'soutstanding_spawns, and an agent that dies mid-generation leaves zero trace while you wait. For the correctsubagent_typeand why a wrong identity is now blocked, seeshared/spawn-identity.md— match the frontmattername: ai-sdlc-planner, never a generic agent. The planner can only write underai/<run>/and.claude/context/(guard-enforced — never repo source), so point it at.claude/context/repo-map/<name>/. It follows the map content contract (${CLAUDE_PLUGIN_ROOT}/skills/dev-workflow/steps/repo-map-task.md):index.md(purpose/stack/modules/cross-repo edges — intake's targeting tier),areas/*detail files (each loadable alone), andconventions.md(observed patterns with cited examples — what plan-review checks plans against). - Stamp it yourself, not the planner:
${CLAUDE_PLUGIN_ROOT}/bin/harness repo-map-stamp --repo-name <n> --repo <path>— stamping is the orchestrator's job, never the planner's own. - Remind the user: the planner still reads real code for areas it plans to touch — the map speeds targeting, it doesn't replace reading.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

