agentleFS
hasuwini77hasuwini77/agentille/skills/agentille-project/SKILL.md

agentille-project

Skill1 starsChanged 20 days agoMITView on GitHub

Register the current repo with agentille and write its ./CLAUDE.md. Asks 7 short per-repo questions, appends the project to ~/.agentille/profile.json's projects[] and renders a per-repo CLAUDE.md using your voice settings. Optional; /agt works without it. Runs only when the user types /agentille-project; model invocation is disabled.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    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:

    Run inside a repo to register it and seed its ./CLAUDE.md.

    line 9
  2. Very specific:

    Profile: If ~/.agentille/profile.json is absent, continue with {} (voice lines are then omitted from the CLAUDE.md).

    line 13
  3. Very specific:

    Existing ./CLAUDE.md: If present, ask: keep it or overwrite?

    line 14
  4. Distinctive:

    Never overwrite a file without explicit confirmation.

    line 36
  5. Distinctive:

    Never change the profile's voice settings.

    line 37

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

agentille-project — per-repo registration

Run inside a repo to register it and seed its ./CLAUDE.md. Never auto-trigger.

Steps

  1. Profile. If ~/.agentille/profile.json is absent, continue with {} (voice lines are then omitted from the CLAUDE.md). If it exists but is not valid JSON, stop and tell the user.
  2. Existing ./CLAUDE.md. If present, ask: keep it or overwrite? Skip the render on keep (still register the project). Never overwrite without an explicit yes.
  3. Ask once, in one message, each with a default:
    1. name (directory basename)
    2. description (one line)
    3. techStack (free-form list; offer what you detect from the repo)
    4. goals
    5. claudeUse — how Claude is used here
    6. constraints — rules Claude must respect
    7. integration — where finished work lands: pr (open a PR), push (push my feature branch, no merge), local (no remote)
  4. Register. Read-merge-write ~/.agentille/profile.json: append to projects[], or replace the entry with the same url or path (else the same id). Touch nothing else in the file. url is the origin remote (git remote get-url origin, omit the key when there is none) and path the repo root: the mod's Bash guard looks a run's integration mode up by the repo a command targets, matched on url or path; an entry with only an id counts for a repo with no remote and can only lower the mode.

    { "id": "<slug>", "url": "<origin remote URL>", "path": "<repo root>", "name": "", "description": "",
      "techStack": [], "goals": "", "claudeUse": "", "constraints": "", "integration": "pr | push | local" }

    Slug: lowercase, non-alphanumeric to -, trimmed; empty → project-<random>.

  5. Render ./CLAUDE.md from claude-md-template.md, pulling name, deliveryStyle, tone, neverDo from the profile.
  6. Write ./.agentille/config.json as { "integration": "<chosen>" } (repo-local; it can only lower the mode your profile sets, never raise it: auto comes from the profile alone). Print the three paths and suggest adding .agentille/ to .gitignore if it should not be committed.

Hard rules

  • Never overwrite a file without explicit confirmation.
  • Never change the profile's voice settings; this skill is project-scoped.
  • Never commit ~/.agentille/profile.json.

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.