agentleFS
majiayu000majiayu000/test-loop/.claude/skills/test-loop-bootstrap/SKILL.md

test-loop-bootstrap

Skill0 starsChanged 37 days agoMITView on GitHub

Meta-skill: choose which of the other five test-loop skills to invoke first, based on the user's description of the current state of their project. Use when a user says something like "I want to add tests to this project" or "this project has no CI" and the right entry point is not obvious.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When a user asks for help with testing, CI, or quality on a project and it is not obvious which of the five concrete skills to start with.

    line 23
  2. Distinctive:

    If you already know the user wants, say, drift detection, call drift-check directly.

    line 26
  3. Distinctive:

    Routing every request through this skill: If the user already named a skill, call it.

    line 87

≈ “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

test-loop-bootstrap

Pick the right next skill. The five other skills are pieces of one loop; this skill is the door.

When to use

When a user asks for help with testing, CI, or quality on a project and it is not obvious which of the five concrete skills to start with.

If you already know the user wants, say, drift detection, call drift-check directly. Do not use this meta-skill as an indirection layer.

Inputs

A free-form description of where the project is now. Examples:

  • "I just started a new Swift project and I want to set up testing."
  • "I have a Python project that already has pytest but no CI."
  • "My existing tests pass but I don't trust them — I want to know they actually run."
  • "Someone added a public function and I want to know if it broke anything."

Decision tree

Q1: Does the project have at least one of:
    - a pre-commit hook calling scripts/check_drift.sh
    - a CI workflow calling swift test / pytest / etc.
    - a docs/knowledge/ directory
    --> If no:        start with [init-loop]
    --> If yes:       continue to Q2

Q2: Did something just change in the public API (a new struct, function,
    class, or trait)?
    --> If yes:       run [drift-check] now
    --> If no:       continue to Q3

Q3: Did the test suite just run, and did it produce at least one
    failure?
    --> If yes:       run [failure-classify], then [report-render]
    --> If no:       continue to Q4

Q4: Have you (or anyone on the team) edited the loop tools
    (scripts/*.sh, scripts/*.py, .githooks/*, .github/workflows/*)
    in the last week?
    --> If yes:       run [drill] to verify the loop still reacts
    --> If no:       continue to Q5

Q5: When is the last time [drill] was run?
    --> > 30 days:    run [drill] as a regression check
    --> <= 30 days:   stop. The loop is healthy. No skill needed.

Output

One of:

  • [init-loop]
  • [drift-check]
  • [failure-classify] + [report-render]
  • [drill]
  • "no skill needed; the loop is healthy"

In addition, the skill should briefly state why it chose what it chose, so the user can override.

Anti-patterns

  • Routing every request through this skill. If the user already named a skill, call it. This skill is for undecided situations.
  • Adding more branches to the decision tree. The five Q&As already cover the realistic states. Adding a sixth tends to increase router complexity without improving routing accuracy.
  • Recommending [drill] on a project that does not yet have the loop. The drill requires the loop to be installed. If the answer to Q1 is "no", go to [init-loop] first.

Cross-references

  • init-loop
  • drift-check
  • failure-classify
  • report-render
  • drill

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.