agentleFS
DanielGodriNetoDanielGodriNeto/claude-skills/skills/project-map/SKILL.md

project-map

Skill1 starsChanged 38 days agoMITView on GitHub

Use when the user asks to explain, understand, map, or get an overview of a project or codebase ("explain this project", "how does this repo work", "map the features"). Builds an Obsidian vault with one overview note per feature plus a graph of the code, so the user can browse the project visually.

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

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

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

    Build the graph (cheap, local AST for code): run the graphify skill on the project root with --obsidian --obsidian-dir …

    line 10

“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

  1. Size gate (tokens): if the project is small (about 30 files or fewer), skip the vault: read the README and key files and answer in a few lines, then offer the map. Run the full flow for non-trivial projects or when the user asks for a map/vault.
  2. Build the graph (cheap, local AST for code): run the graphify skill on the project root with --obsidian --obsidian-dir ~/vaults/<project-name>. Skip if ~/vaults/<project-name> is newer than the last commit.
  3. Write ~/vaults/<project-name>/00-Overview.md by hand from graphify-out/GRAPH_REPORT.md and the code:
    • 3-line purpose of the project and its stack.
    • One section per feature (not per file): plain-language explanation in 2-4 lines, the key files, and [[wikilinks]] to the matching graphify node notes.
    • "How the pieces connect" (short), "Risks / rough edges", "Where to start reading". 2b. Two variants (public-safe). The full vault stays private. Judge sensitivity: if any feature, endpoint, config, auth/payment/anti-abuse logic or internal detail could hurt the project in the wrong hands (someone wanting to sabotage it), build ~/vaults/<project-name>-public that omits or generalizes those parts, and tell the user in chat what was left out (never write that list inside the public vault). If nothing is dangerous, say so: the public variant equals the full one. Only the public variant may go to a public repo.
  4. Reply in the chat in 5-8 lines: what the project is, the 3-5 main features, and the path of the vault. Tell the user to open that folder as a vault in Obsidian and use Graph View; invite questions. Do not paste the whole overview.
  5. Offer to commit it (ask once per project): in the same repo under docs/project-map/ or in a separate repo, private or public. Public repos get only the public variant, scanned for secrets, tokens and internal paths first. Follow the global Git rules (commit locally; auto-push only if the repo is private and the change is small), and always ask before pushing it to a public repo.

Rules: keep notes factual (read the code, don't guess); build the vault only under ~/vaults, and copy it into the repo (docs/project-map/) only if the user chooses to commit it; if the user has an existing vault, ask once for its path and reuse it.

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.