AD

Provides a structured method for adding new lint, test, or security checks into the CI pipeline.

Install

mkdir -p .claude/skills/add-ci-gate && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/18125" && unzip -o skill.zip -d .claude/skills/add-ci-gate && rm skill.zip

Installs to .claude/skills/add-ci-gate

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

The discipline for adding or changing a check in Murmur's CI gate the right way — put it in scripts/ci.sh (the single source of truth), in the correct order, tool-absent-safe, without slowing the iterate loop, and (for a guardrail hook) with a RED-before-GREEN selftest assertion; then confirm GitHub Actions still just wraps ci.sh. Use whenever the user wants to add a lint/test/security/format/build check to CI, tighten the gate, or wire a new tool into the pipeline.
470 charsno explicit “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)
Advanced

Key capabilities

  • →Add new checks to scripts/ci.sh
  • →Ensure checks are tool-absent-safe
  • →Prevent slowing down the inner iterate loop
  • →Prove guardrail hooks with RED-before-GREEN assertions
  • →Verify runner prerequisites in .github/workflows/ci.yml

How it works

The skill guides the addition of CI checks by modifying scripts/ci.sh, ensuring tool-absent safety and proper ordering. It also includes steps for proving guardrail hooks and verifying runner prerequisites.

Inputs & outputs

You give it
User request to add a lint/test/security/format/build check to CI, tighten the gate, or wire a new tool into the pipeline
You get back
Modified scripts/ci.sh, potentially modified .github/workflows/ci.yml, and verification of the new CI gate

When to use add-ci-gate

  • →Add new linting rules
  • →Integrate security scanning
  • →Add custom build checks
  • →Define pipeline guardrails

About add-ci-gate

Enforces the discipline of adding CI gates to the source-of-truth script rather than configuration files. It ensures checks are efficient, tool-safe, and verified by selftests.

The discipline for adding or changing a check in Murmur's CI gate the right way — put it in scripts/ci.sh (the single source of truth), in the correct order, tool-absent-safe, without slowing the iterate loop, and (for a guardrail hook) with a RED-before-GREEN selftest assertion; then confirm GitHub

When not to use it

  • →Adding checks directly to .github/workflows/ci.yml for non-prerequisite steps
  • →Wiring expensive checks into anything that runs every edit
  • →Adding a bare cargo clippy --all-targets anywhere an agent would run it directly

Limitations

  • →Does not allow adding checks directly to .github/workflows/ci.yml unless it's a runner-level prerequisite
  • →Does not allow adding expensive checks to the inner iterate loop
  • →Does not allow bare cargo clippy --all-targets outside of ci.sh

How it compares

This skill enforces a specific discipline for adding CI gates by centralizing checks in a single script and verifying them with RED-before-GREEN tests, preventing CI drift and ensuring consistency between local and cloud gates.

Compared to similar skills

add-ci-gate side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
add-ci-gate (this skill)03moReviewAdvanced
ci-cd09moReviewIntermediate
cicd-smoke07moNo flagsIntermediate
fixing-pipeline-errors07moReviewIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry