agentleFS
Sign inStart free
hmj1026hmj1026/dhpk/skills/repo-verify/SKILL.md

repo-verify

Skill2 starsChanged 8 days agoMITView on GitHub

Run a repository verification loop with runner-first execution, deterministic stages, and explicit PASS/FAIL/SKIP output. Use for fast or full validation of a consumer repository. Not for: mutation, fixing failures, dependency remediation, or merge approval. Output: runner summary or a complete fallback stage table.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    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:

    Run read-only in the consumer repository.

    line 21

≈ “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

Repository Verify

Workflow

Run read-only in the consumer repository. Normalize no mode or full to full stages and fast to lint + unit; accept only fast or full. Attach --integration <path> and --e2e <path> only to their matching stages.

Use a project-installed .claude/dhpk/skills/repo-verify/scripts/verify-runner.js when present. If it is absent, use the package-local $SKILL_DIR/scripts/verify-runner.js for a Node project. $SKILL_DIR is the installed directory containing this SKILL.md; its adjacent scripts/lib/runner-utils.js is part of the same local closure, and the consumer cwd remains unchanged. A runner's crash, missing overall verdict, execution error, or explicit semantic FAIL is terminal and does not fall through. A Markdown FAIL or summary.json overallPass=false remains failure even when the runner exits zero. When no applicable runner is available, use the non-Node/manifest fallback and stage rules in references/workflow.md; unsupported manifests are terminal failures.

For evidence applicability and reuse, follow the kernel's Evidence snapshots rule.

When NOT to Use

  • A check should edit, format, build, or fix the repository.
  • Dependency security review is the goal (use $change-verdict deps).
  • Merge approval or code-review verdict is the goal.

Output

For a runner, prefix output with Source: runner and preserve its summary. For fallback, report Source: fallback, every lint/typecheck/unit/ integration/e2e stage as PASS, FAIL, or SKIP with command/reason, and an agreeing overall verdict. Invalid modes, unsupported manifests, runner failures, and execution errors are terminal FAIL/BLOCKED results.

Verification

  • Mode and optional paths were normalized before execution.
  • Runner precedence and terminal failure boundary were respected.
  • Every applicable stage has a result and exact command or skip reason.
  • Semantic overall verdict agrees with the stage results and source is named, regardless of CLI exit status.

References

  • references/workflow.md — runner precedence, fallback ecosystems, Node script rules, and report template.

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