agentleFS
TidalPaladinTidalPaladin/skills/standardize-ci/SKILL.md

standardize-ci

Skill1 starsChanged 24 days agoApache-2.0View on GitHub

Plan CI initialization, restructuring, or runtime reduction without changing files or remote state.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    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:

    Keep this skill planning-only in every mode.

    line 8
  2. Distinctive:

    Use the supplied repository or the current one.

    line 12
  3. Distinctive:

    Read relevant workflow files, manifests, quality commands, supported runtimes, and release requirements.

    line 17
  4. Distinctive:

    Do not require confirmation of facts already established by the repository.

    line 24
  5. Distinctive:

    Keep fast, useful feedback on pull requests.

    line 35
  6. Distinctive:

    Keep untrusted fork code off persistent self-hosted runners.

    line 40
  7. Distinctive:

    Ask about trust or capacity only when the available policy does not resolve it.

    line 42
  8. Distinctive:

    Keep required-check transitions explicit.

    line 49
  9. Distinctive:

    Use existing schedules when suitable.

    line 51

≈ “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

Standardize CI

Keep this skill planning-only in every mode. Inspect the target and return a decision-complete plan. Do not edit files, register runners, dispatch workflows, change settings, or publish branches.

Use the supplied repository or the current one. Honor an existing provider or an explicit choice. For new CI with no provider selected, default to GitHub Actions.

Inspect and Decide

Read relevant workflow files, manifests, quality commands, supported runtimes, and release requirements. Trace job commands through scripts and package hooks to identify required tools and services. Use available run history for duration, queue, setup, cache, and artifact evidence.

Reuse established policy and facts. Ask only for unresolved decisions that affect cost, supported platforms, runner trust, required checks, artifacts, or schedules. Do not require confirmation of facts already established by the repository.

For GitHub Actions, use the relevant sections of planning patterns:

  • Job allocation and runtime controls for workflow splits or latency work.
  • Dependency health for advisory and deprecation checks.
  • Runner trust and bootstrap for self-hosted execution.
  • Security for permissions, action pins, and cache boundaries.
  • Scheduled validation for new or changed schedules.

Keep fast, useful feedback on pull requests. Allocate costly jobs from measured runtime, capacity, and support requirements. Do not invent artifacts or add release builds to repositories that have no production-build requirement. Preserve declared runtime and critical-library boundary coverage.

Keep untrusted fork code off persistent self-hosted runners. Define hosted fallbacks or conclusive aggregate checks where branch protection requires them. Ask about trust or capacity only when the available policy does not resolve it.

Plan and Validation

Specify the workflow and job names, commands, runners, setup, pins, triggers, permissions, caches, artifacts, timeouts, concurrency, and dependencies that apply. For self-hosted jobs, include bootstrap exceptions and early preflight failures. Keep required-check transitions explicit.

Use existing schedules when suitable. For new schedules, settle operational constraints with the user when they matter. Otherwise choose a non-zero UTC minute and state the proposed cadence and time as defaults.

For new or changed scheduled workflows, include exact-ref dispatch evidence under GitHub Actions validation. Honor its authorization and notification thresholds.

Return one <proposed_plan> block with the changes, job table, validation, required-check transition, and assumptions. Stop when implementation needs no unresolved material decision.

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.