agentleFS
omarismailbomarismailb/product-wiki/.agents/skills/generate-checks/SKILL.md

generate-checks

Skill8 starsChanged 5 months agoMITView on GitHub

Use when product wiki acceptance criteria, rules, journeys, or outcomes need executable checks, regression tests, evaluations, or verification evidence.

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. Distinctive:

    Read the relevant wiki units.

    line 20
  2. Very specific:

    Add a PW: anchor in the test when it helps the agent jump from the wiki claim to the check.

    line 36
  3. Very specific:

    Update checks/manifest.json when the check should remain part of the product contract.

    line 37
  4. Very specific:

    Run node scripts/checks-lint.mjs --run and node scripts/wiki-anchor-lint.mjs.

    line 39

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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

Generate Checks

Your job is to turn product wiki claims into checks that run against code.

Inputs

  • Acceptance criteria.
  • Product rules.
  • Journeys or flows.
  • Outcomes when measurable.
  • Existing test conventions.

Workflow

  1. Read the relevant wiki units.
  2. Read references/check-patterns.md.
  3. Locate current test patterns.
  4. Choose the check type:
    • unit
    • integration
    • E2E
    • lint
    • typecheck
    • eval
    • manual evidence
  5. Apply the edge-case triggers in references/check-patterns.md. For each criterion, add a check for every trigger it matches, not only the happy path, and record any matched trigger you deliberately leave unchecked.
  6. For behaviour changes, write the smallest check first and run it before implementation. Confirm it fails for the expected reason when the behaviour is not yet implemented.
  7. Link the check back to the wiki unit by stable ID. Add a PW: anchor in the test when it helps the agent jump from the wiki claim to the check.
  8. Update checks/manifest.json when the check should remain part of the product contract.
  9. Run the check.
  10. Run node scripts/checks-lint.mjs --run and node scripts/wiki-anchor-lint.mjs.

Output

Return:

  • Wiki IDs covered.
  • Checks added.
  • PW: anchors added where useful.
  • Command run.
  • Result.
  • Whether the check was observed failing before implementation, when applicable.
  • Edge-case triggers each criterion matched, and which check covers each (or why a match is deliberately left unchecked).
  • Any acceptance criteria still uncovered.

Related skills

  • compile-change: when checks are part of an implementation.
  • reconcile-wiki: when finding missing coverage after implementation.

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.