agentleFS
PMDevSolutionsPMDevSolutions/Aurelius/.claude/skills/ci/SKILL.md

ci

Skill9 starsChanged 19 days agoMITView on GitHub

Run the same checks as /verify-all non-interactively, with JSON output and an exit code that reflects overall pass or fail, for CI pipelines, pre-merge gates and automation. Keywords: ci, json, pre-merge, automation, verify

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When invoked from a CI pipeline, pipe the JSON straight to a downstream step or store it as a build artifact.

    line 77

≈ “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

/ci — CI-Optimized Verification

Runs the same checks as /verify-all but in non-interactive, machine-readable mode. Designed for CI pipelines, pre-merge gates, and automation: no prompts, JSON output, exit code reflects overall pass/fail.

Usage

/ci
/ci --skip dead-code,security
/ci --include lint-and-format,types,tests

The slash command wraps ./scripts/verify-all.sh --ci. The --ci flag implies --json and sets non-interactive behavior. Pass --skip or --include to filter checks.

What runs

Same eleven checks as /verify-all, in the same order:

lint-and-format → types → tests → accessibility → tokens → dead-code → security → bundle-size → agent-plugins → renderers → claude-code

bundle-size is automatically skipped when no build artifact (dist/, .next/, build/, out/) is present, and agent-plugins when .claude/agent-plugins/ has none. No check requires a dev server, so all of them run in CI without extra setup.

Output

JSON written to stdout:

{
  "ok": false,
  "summary": { "pass": 8, "fail": 1, "skip": 2, "total": 11 },
  "checks": [
    { "name": "lint-and-format", "status": "pass", "exitCode": 0, "durationMs": 2618, "reason": "" },
    { "name": "types",           "status": "fail", "exitCode": 1, "durationMs": 1843, "reason": "exit 1" },
    { "name": "tests",           "status": "pass", "exitCode": 0, "durationMs": 4205, "reason": "" },
    ...
    { "name": "bundle-size",     "status": "skip", "exitCode": 0, "durationMs": 0,    "reason": "no build artifact (dist/.next/build/out)" }
  ]
}
FieldMeaning
oktrue only when every check passed or was skipped intentionally
summary.pass / fail / skip / totalCounts across all checks
checks[].statuspass | fail | skip
checks[].exitCodeThe exit code returned by the underlying check script
checks[].durationMsWall-clock duration
checks[].reasonEmpty on pass; short reason on fail/skip

Steps

1. Run the orchestrator in CI mode

./scripts/verify-all.sh --ci $ARGUMENTS

2. Surface results to the user

Parse the JSON and report the summary in plain text:

8 passed, 1 failed, 2 skipped
Failing: types (exit 1, 1843ms)

When invoked from a CI pipeline, pipe the JSON straight to a downstream step or store it as a build artifact.

Exit codes

  • 0 — every check passed or was skipped
  • 1 — one or more checks failed

The exit code is what CI runners pivot on, so do not swallow it.

Differences vs /verify-all

Aspect/verify-all/ci
OutputHuman-readable summary tableJSON to stdout
InteractivityAllowed (no prompts today, but reserved)Forbidden
Use caseLocal developmentCI pipelines, automation, scripts
Exit code on failure11

Related

  • /verify-all — human-readable version of the same checks
  • ./scripts/verify-all.sh --help — full flag documentation

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.