agentleFS
blixvipblixvip/skills/github/SKILL.md

github

Skill4 starsChanged 3 months agoMITView on GitHub

Quickly audit, document, validate, commit, and safely publish or update a project on GitHub. Use for /github, publishing, repository setup or updates, README or repository-health work, pre-publish checks, and launch reviews. Supports private, public, update, docs-only, no-push, and launch.

166 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use taskflow-inbox-triage, BlueBubbles and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • AtomicBot-aiSkill

    taskflow-inbox-triage

    Also in 6 projects that use this

  • jiulingyunSkill

    BlueBubbles

    为Clawdbot构建或更新BlueBubbles外部通道插件(扩展包,REST发送/探测,webhook入站)。

    Also in 20 projects that use this

  • hkyutongSkill

    node-connect

    Diagnose OpenClaw node connection and pairing failures for Android, iOS, and macOS companion apps. Use when QR/setup code/manual connect fails, local Wi-Fi works but VPS/tailnet does not, or errors mention pairing required, unauthorized, bootstrap token invalid or expired, gateway.bind, gateway.remote.url, Tailscale, or plugins.entries.device-pair.config.publicUrl.

    Also in 18 projects that use this

  • molt-botSkill

    discord-e2e

    Test Discord with Convex-leased QA bots: check credentials without a Gateway, exercise owned messages, or prove OpenClaw replies, files, threads, reactions, streaming, and typing. Distinguish API and Gateway evidence from manual user interactions.

    Also in 10 projects that use this

  • molt-botSkill

    openclaw-repair-sweep

    Run scoped OpenClaw issue/PR repair campaigns: coordinate workers, prove root causes, and land or close verified work under the requested authority.

    Also in 10 projects that use this

  • clawdbotSkill

    openclaw-performance

    Performance work in OpenClaw: a slow or heavy flow (latency, CPU, memory, SQL, RPC, renders), writing or reviewing a perf PR, or adding a cache. Count the work, delete waste first, cache last.

    Also in 7 projects that use this

  • +2 more from molt-bot

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 151 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    launch: Run a comprehensive public-launch review.

    line 17
  2. Very specific:

    Run configured lint, test, type-check, format-check, and build commands in parallel when safe.

    line 27
  3. Distinctive:

    Never push secrets, private data, accidental generated or dependency files, or files over GitHub's limit.

    line 34
  4. Distinctive:

    Never use git add ., force-push, delete repositories or branches, rewrite history, bypass protection, or change an existing …

    line 35
  5. Distinctive:

    Never fabricate project claims, badges, compatibility, metrics, demos, licenses, or policies.

    line 36
  6. Distinctive:

    Default new repositories to private and preserve existing visibility.

    line 37
  7. Distinctive:

    Read references/safety.md only when the audit finds a risk.

    line 40

“Common” is measured: kept by at least 6 in 10 of the 151 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

Fast GitHub Publisher

Finish as quickly as possible without weakening safety or making unverified claims. Start work immediately, batch independent reads and checks, make the fewest justified edits, and avoid optional ceremony. Keep updates and the final report terse.

Options

  • Default: Prepare and publish; preserve existing visibility, or make a new repository private.
  • private / public: Set new-repository visibility. Confirm before changing an existing private repository to public.
  • update: Include intended current changes, validate, commit, and push.
  • docs-only: Edit documentation only; never push.
  • no-push: Prepare and commit locally; never mutate the remote.
  • launch: Run a comprehensive public-launch review.

Combine compatible options. docs-only and no-push disable pushing.

Fast workflow

  1. Snapshot Git status, branch, remotes, and recent commits. Run python <skill-dir>/scripts/project_audit.py --path . --format text.
  2. Inspect only the audit-identified manifests, commands, entry points, configuration, tests, and docs needed to understand and describe the project. Run independent reads and checks in parallel.
  3. Resolve safety blockers. Preserve unrelated user changes. Never expose secret values.
  4. Make minimal evidence-based edits. Read references/readme.md only when changing README.md; read references/templates.md only when adding repository files; read references/research.md only for launch.
  5. Run configured lint, test, type-check, format-check, and build commands in parallel when safe. Do not invent checks or reinstall dependencies unnecessarily. Re-run affected failures, then run the audit with --strict.
  6. Review the diff and stage explicit paths. Commit logically without rewriting history.
  7. For remote work, read references/github-operations.md, check active gh authentication, run remote_guard.py in dry-run mode, execute it with --push, then verify the remote SHA and metadata.
  8. Report URL, changed files, checks or failures, commits, remote actions, and blockers—nothing extra.

Non-negotiable safety

  • Never push secrets, private data, accidental generated or dependency files, or files over GitHub's limit. If a credential may be committed, stop and recommend rotation plus history remediation.
  • Never use git add ., force-push, delete repositories or branches, rewrite history, bypass protection, or change an existing visibility/default branch without explicit authorization.
  • Never fabricate project claims, badges, compatibility, metrics, demos, licenses, or policies.
  • Default new repositories to private and preserve existing visibility.
  • A failed check that makes documentation false blocks publishing. Report unrelated failures without expanding scope.

Read references/safety.md only when the audit finds a risk. Read references/workflow.md for ambiguous or complex repositories, references/usage-and-troubleshooting.md for failures, and references/github-operations.md for remote actions. Use the bundled scripts as the preferred fast path.

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.