agentleFS
sdwurg180507280211sdwurg180507280211/my-skills/skills/github-kb/SKILL.md

github-kb

Skill1 starsChanged 38 days agoMITView on GitHub

Maintain and search a local knowledge base of cloned GitHub repositories. Use when the user wants to search local repositories, reuse local source code, clone a repository into the local project collection, or maintain a local repo registry. Do not trigger for ordinary public GitHub questions that do not need the local repository collection.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 8 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    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. Distinctive:

    Use a local repository collection as the first source when the task is specifically about the user's cloned projects or reusable local code.

    line 8
  2. Very specific:

    Prefer an explicit project root supplied by the user.

    line 12
  3. Distinctive:

    Check the local repository root for an exact or likely repository match.

    line 16
  4. Distinctive:

    If a local repository exists, inspect it locally before searching remote GitHub.

    line 17
  5. Distinctive:

    If the requested repository is not local and the user wants a local copy, clone it into the repository root.

    line 18
  6. Distinctive:

    Never overwrite an existing local repository silently.

    line 20
  7. Very specific:

    Use the environment's native GitHub connector/tool when available.

    line 33
  8. Distinctive:

    Do not trigger on every mention of “GitHub”, “repo”, or “repository”.

    line 37
  9. Distinctive:

    Do not clone repositories unless a local copy is useful for the requested task.

    line 38
  10. Very specific:

    Do not assume ~/IdeaProjects or another machine-specific path.

    line 39
  11. Distinctive:

    Do not modify the user's local registry or clone location without a clear task reason.

    line 40

≈ “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

GitHub Local Knowledge Base

Use a local repository collection as the first source when the task is specifically about the user's cloned projects or reusable local code.

Repository root

Prefer an explicit project root supplied by the user. Otherwise use GITHUB_KB_ROOT when set; if neither is available, use ~/projects as a conventional default and confirm it exists before writing there.

Workflow

  1. Check the local repository root for an exact or likely repository match.
  2. If a local repository exists, inspect it locally before searching remote GitHub.
  3. If the requested repository is not local and the user wants a local copy, clone it into the repository root.
  4. Maintain an optional CLAUDE.md registry in the repository root with one concise line per cloned repository.
  5. Never overwrite an existing local repository silently.

Local registry format

# GitHub Repository Registry

- **repo-name**: One-line description
- **another-repo**: One-line description

GitHub access

Use the environment's native GitHub connector/tool when available. Fall back to gh or the GitHub REST API only when appropriate for the runtime.

Guardrails

  • Do not trigger on every mention of “GitHub”, “repo”, or “repository”.
  • Do not clone repositories unless a local copy is useful for the requested task.
  • Do not assume ~/IdeaProjects or another machine-specific path.
  • Do not modify the user's local registry or clone location without a clear task reason.

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.