agentleFS
Sign inStart free
ThalisuThalisu/SKILLS/skills/discover-setup/SKILL.md

discover-setup

Skill1 starsChanged 43 days agoMITView on GitHub

Install or refresh discover on this machine and in one CLAUDE.md, the project's or the user's global one: links the discover agent and the two discover skills, then installs or updates the mandatory Discovery section.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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 bash <this-skill-dir>/scripts/install.sh <scope-args>.

    line 22
  2. Very specific:

    Run bash <this-skill-dir>/scripts/verify.sh <scope-args> again.

    line 28
  3. Very specific:

    Only if this session was prompted for permission on the agent's Bash call (bash ~/.claude/skills/discover/scripts/discover.sh): offer, and …

    line 30
  4. Distinctive:

    Do not commit: a project CLAUDE.md change is left for the user to review.

    line 35

≈ “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

Install or refresh discover. Runs inline: it edits a CLAUDE.md, which the forked discover agent cannot do. Every script below takes --user for the user scope (~/.claude/CLAUDE.md) and a project directory for the project scope (<project>/CLAUDE.md).

Scope argument: $ARGUMENTS

  1. Pick the scope(s). If the argument is project, user or both, use it and skip the question. Otherwise run bash <this-skill-dir>/scripts/verify.sh <project> and bash <this-skill-dir>/scripts/verify.sh --user, then ask with AskUserQuestion (multiSelect, so both may be chosen), showing each scope's section= state in its description:
    • Project: <project>/CLAUDE.md; shared with teammates through the repo.
    • User: ~/.claude/CLAUDE.md; applies to every project on this machine, yours only. Recommend the one that is not current yet; when both are current, say so and offer only a refresh.
  2. For each chosen scope, run bash <this-skill-dir>/scripts/verify.sh <scope-args> and read section=, agent_link=, skill_links=, deps=. A deps entry :missing blocks the install: say which tool is missing (rg, ast-grep, jq) and stop.
  3. Run bash <this-skill-dir>/scripts/install.sh <scope-args>. It creates or repairs the agent symlink and the two skill hops, then handles the section: none → appends it at the end of that CLAUDE.md; stale (an older version installed) → updates the text between the markers; current → no change. On drifted it prints the diff and exits 5: show the diff, ask whether the hand edits may be replaced, and only then re-run with --force. The script touches nothing else in that CLAUDE.md, and nothing else in the project or in ~/.claude.
  4. Run bash <this-skill-dir>/scripts/verify.sh <scope-args> again; it must exit 0. Report, per scope, the transition (section: none -> current v=3, stale v=2 -> current v=3), the link states and agent_ref.
  5. Only if this session was prompted for permission on the agent's Bash call (bash ~/.claude/skills/discover/scripts/discover.sh): offer, and wait for a yes, to add Bash(bash ~/.claude/skills/discover/scripts/discover.sh:*) to permissions.allow in ~/.claude/settings.json. Never edit that file without the explicit yes.

Do not commit: a project CLAUDE.md change is left for the user to review.

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.

Check your setup, free