agentleFS
egregore-labsegregore-labs/egregore/.claude/skills/test/SKILL.md

test

Skill293 starsChanged 10 days agoMITView on GitHub

Run Egregore's fast static, Runtime-contract, and targeted automated test gate before save. Optional relationship index validation requires an explicit request.

185 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use agent-context-isolation, agentica-infrastructure 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.

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 158 edited versions we read keep these.

  • 8

    Distinctive

    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:

    Run the fast automated gate for the current change.

    line 6
  2. Very specific:

    Use for $test, “test my changes”, or a quick pre-save validation.

    line 10
  3. Very specific:

    Run bash bin/base-branch.sh to print the configured integration branch.

    line 17
  4. Distinctive:

    Run static analysis once:

    line 19
  5. Distinctive:

    Run the architectural ratchet once (source instances only — installed instances do not ship the distribution engine and skip this step):

    line 28
  6. Very specific:

    Only when the user explicitly passed --graph or changed the graph projection adapter itself:

    line 49
  7. Distinctive:

    Never scan every skill for Cypher and execute it against a live database.

    line 58
  8. Distinctive:

    No default Neo4j, API, or network dependency.

    line 76
  9. Distinctive:

    Use isolated fixtures for writes and destructive/failure behavior.

    line 79
  10. Distinctive:

    Never call graph functionality the correct default Runtime boundary.

    line 80
  11. Distinctive:

    Do not claim “safe to save” when required suites were skipped.

    line 81

“Common” is measured: kept by at least 6 in 10 of the 158 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

Run the fast automated gate for the current change.

When to invoke

Use for $test, “test my changes”, or a quick pre-save validation. Use qa for broader behavioral proof and adversarial fixtures.

Arguments: $ARGUMENTS — optional paths, --all, or explicit --graph.

Default gate

  1. Run bash bin/base-branch.sh to print the configured integration branch. Scope to supplied paths, --all, or changed files against that branch.
  2. Run static analysis once:
bash bin/test-changes.sh <scope>

--all scans source skills plus Runtime-neutral shell adapters. Do not treat a missing historical .claude/commands directory as test coverage.

  1. Run the architectural ratchet once (source instances only — installed instances do not ship the distribution engine and skip this step):
if [ -f bin/capability-distribution.mjs ] && [ -f capability-distribution.json ]; then
  bash bin/node-run.sh bin/capability-distribution.mjs runtime-skill-audit \
    --root . --strict-contract
fi

This checks the complete skill denominator, QMD/graph/control-plane/write coupling ceilings, and migrated-skill contracts.

  1. Select the smallest relevant existing test suites from the changed paths and skill-references.json. Run independent suites in parallel where the harness supports it. Do not rerun the same suite through multiple rituals.
  2. For changes to cross-harness Runtime sources or skills, run Codex/Pi/Prime bundle parity checks.

Optional graph validation

Only when the user explicitly passed --graph or changed the graph projection adapter itself:

  • require Connected graph capability;
  • extract bounded read-only Cypher fixtures from the named graph files only;
  • reject CREATE, MERGE, SET, DELETE, REMOVE, or DROP;
  • validate through the optional graph projection adapter;
  • report graph unavailable as an optional-check gap, not a Local test failure.

Never scan every skill for Cypher and execute it against a live database.

Result

Render one compact pass/fail report with:

  • static checks;
  • Runtime architecture audit;
  • targeted suites and counts;
  • bundle parity when applicable;
  • optional graph check only when requested;
  • exact failures and the safe next action.

Warnings do not block unless they represent a privacy, authorization, canonical-order, or destructive-action risk.

Rules

  • No default Neo4j, API, or network dependency.
  • Tests must not modify production/live QMD, graph, control-plane, or canonical memory state.
  • Use isolated fixtures for writes and destructive/failure behavior.
  • Never call graph functionality the correct default Runtime boundary.
  • Do not claim “safe to save” when required suites were skipped.

Emit content-free telemetry after the gate:

bash bin/telemetry.sh emit "command" '{"command":"test"}' 2>/dev/null &

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.