agentleFS
TheophilusChinomonaTheophilusChinomona/idev/skills/project-map/SKILL.md

project-map

Skill0 starsChanged 5 months agoMITView on GitHub

Generates and maintains a detailed project map (all FE/BE files with mappings) at .claude/idev/project-map/project.map.md via Python watcher scripts. Use when generating or refreshing the project map, configuring the map watcher, or when another skill needs the map as a data source.

7 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Always access the map via Grep — never load the whole file.

    line 36
  2. Distinctive:

    If the map is missing or stale, regenerate it (command above) before trusting it.

    line 38
  3. Very specific:

    Follow ${CLAUDE_PLUGIN_ROOT}/skills/project-map/map.rules.md for the full rules.

    line 39
  4. Very specific:

    When regenerating, update the Generated: timestamp in the output.

    line 53

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Project Map Skill

Purpose

Maintain a detailed, greppable map of the project — every frontend file (pages, containers, services) and every backend file (controllers, services). The map is the data source for the smart-context, file-index, and api-contract-validation skills.

State (per project)

FilePurpose
.claude/idev/project-map/project.map.mdThe generated map (note the filename: project.map.md, NOT project-map.md)
.claude/idev/project-map/watcher_config.jsonFE/BE paths + project type for the watcher

Scripts (in this skill folder)

ScriptPurpose
${CLAUDE_PLUGIN_ROOT}/skills/project-map/ai_map_updater.pyOne-shot map generation (CLI + importable create_project_map)
${CLAUDE_PLUGIN_ROOT}/skills/project-map/map_watcher.pyLong-running watcher; regenerates the map only when source files change (run from the project root)

Generate the map

# from the project root — auto-detects split vs unified, safe non-interactively
python3 "${CLAUDE_PLUGIN_ROOT}/skills/project-map/ai_map_updater.py"

# or with explicit options
python3 "${CLAUDE_PLUGIN_ROOT}/skills/project-map/ai_map_updater.py" \
  --root . --mode split --frontend ./frontend --backend ./backend

# or start the watcher (interactive config on first run)
python3 "${CLAUDE_PLUGIN_ROOT}/skills/project-map/map_watcher.py"
  • Always access the map via Grep — never load the whole file. Grep for the feature or file name (e.g. Grep "Order" .claude/idev/project-map/project.map.md), then Read only the source files it points to.
  • Load only the relevant section (frontend / backend / a single category) when a section-level view is needed.
  • If the map is missing or stale, regenerate it (command above) before trusting it.
  • Follow ${CLAUDE_PLUGIN_ROOT}/skills/project-map/map.rules.md for the full rules.

Staleness Detection

The map file records a Generated: timestamp at the top. Check staleness:

1. Read the first 5 lines of project.map.md
2. Extract the "Generated: YYYY-MM-DD" date
3. If older than 14 days → warn: "project-map is N days old — results may be stale"
4. If older than 30 days → fail loudly: "project-map is severely stale — regenerate before trusting"
5. If missing → generate before proceeding

The session-start hook also checks file mtime and warns if > 14 days old. When regenerating, update the Generated: timestamp in the output.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.