Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Work with GitHub Actions CI/CD workflows - read live syntax, monitor runs, and debug failures. Use when writing, running, or debugging GitHub Actions workflows.
.claude/skills/mkurman-github-workflows/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -12% | 0% |
Path: scripts/ci_monitor.cjs
bashnode scripts/ci_monitor.cjs <command>
Before using any command:
--help to discover available argumentsRouting Table:
| When You Need | Command | |---------------|---------| | List recent runs | runs [--branch <name>] | | Monitor running workflow | watch <run-id> | | Fail fast in scripts | fail-fast <run-id> | | See why run failed | log-failed <run-id> | | Test pass/fail counts | test-summary <run-id> | | Check action versions | check-actions [file] | | Search logs | grep <run-id> --pattern <regex> | | Wait for deployment | wait-for <run-id> <job> --keyword <text> |
Base URL: https://docs.github.com/en/actions/reference/workflows-and-actions/
Before writing any workflow syntax:
.md file from the URL above| Task | File | Section | |------|------|---------| | Create workflow | workflow-syntax.md | name, on, jobs | | Set triggers | workflow-syntax.md | on | | Set permissions | workflow-syntax.md | permissions | | Concurrency | workflow-syntax.md | concurrency | | Reusable workflow | workflow-syntax.md | on.workflow_call | | Annotations | workflow-commands.md | "Setting an error/warning/notice message" | | Output variables | workflow-commands.md | "Environment files" | | Conditionals | expressions.md | "Operators", "Functions" | | Contexts | contexts.md | "<context> context" | | Events | events-that-trigger-workflows.md | Event tables |
| What | Where | |------|-------| | Action versions | node ci_monitor.cjs check-actions <file> | | Node.js LTS | curl -s https://nodejs.org/dist/index.json \| jq '.[0].version' |
"No errors" is not validation. Prove observable change:
BEFORE: [specific state]
AFTER: [different state]
EVIDENCE: [output from ci_monitor.cjs]references/gh/SKILL.md — gh CLI reference| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-20 | pass→pass | 4,057 | 7,339 | +81% | 1 | 1 | 0% | 630 | 1,007 | +60% | 0 | 0 | — |
case-01 | fail→fail | 5,505 | 4,765 | -13% | 1 | 1 | 0% | 908 | 821 | -10% | 0 | 0 | — |
case-02 | fail→fail | 3,132 | 4,088 | +31% | 1 | 1 | 0% | 390 | 791 | +103% | 0 | 0 | — |
case-03 | fail→fail | 13,711 | 5,075 | -63% | 1 | 1 | 0% | 1,623 | 790 | -51% | 0 | 0 | — |
case-04 | fail→pass | 7,924 | 2,169 | -73% | 1 | 1 | 0% | 1,294 | 928 | -28% | 0 | 0 | — |
case-05 | fail→pass | 11,320 | 2,839 | -75% | 1 | 1 | 0% | 1,973 | 1,149 | -42% | 0 | 0 | — |
case-06 | fail→pass | 7,685 | 3,124 | -59% | 1 | 1 | 0% | 1,377 | 1,152 | -16% | 0 | 0 | — |
case-07 | fail→pass | 10,009 | 3,300 | -67% | 1 | 1 | 0% | 1,650 | 1,177 | -29% | 0 | 0 | — |
case-08 | fail→pass | 5,665 | 1,898 | -66% | 1 | 1 | 0% | 961 | 844 | -12% | 0 | 0 | — |
case-09 | fail→pass | 8,687 | 4,799 | -45% | 1 | 1 | 0% | 1,639 | 1,416 | -14% | 0 | 0 | — |
case-10 | fail→pass | 5,619 | 2,394 | -57% | 1 | 1 | 0% | 980 | 995 | +2% | 0 | 0 | — |
case-11 | pass→pass | 7,937 | 2,399 | -70% | 1 | 1 | 0% | 1,369 | 1,075 | -21% | 0 | 0 | — |
case-12 | pass→pass | 5,253 | 5,063 | -4% | 1 | 1 | 0% | 788 | 1,544 | +96% | 0 | 0 | — |
case-13 | fail→pass | 6,003 | 4,229 | -30% | 1 | 1 | 0% | 928 | 1,312 | +41% | 0 | 0 | — |
case-14 | fail→pass | 7,608 | 2,320 | -70% | 1 | 1 | 0% | 1,267 | 972 | -23% | 0 | 0 | — |
case-19 | pass→pass | 3,927 | 1,977 | -50% | 1 | 1 | 0% | 616 | 914 | +48% | 0 | 0 | — |
case-15 | pass→pass | 6,278 | 3,312 | -47% | 1 | 1 | 0% | 1,193 | 1,114 | -7% | 0 | 0 | — |
case-16 | pass→pass | 5,786 | 1,991 | -66% | 1 | 1 | 0% | 918 | 892 | -3% | 0 | 0 | — |
case-17 | fail→pass | 10,524 | 1,923 | -82% | 1 | 1 | 0% | 1,773 | 884 | -50% | 0 | 0 | — |
case-18 | pass→pass | 3,242 | 2,610 | -19% | 1 | 1 | 0% | 507 | 1,040 | +105% | 0 | 0 | — |
case-21 | pass→pass | 7,597 | 5,337 | -30% | 1 | 1 | 0% | 1,384 | 1,592 | +15% | 0 | 0 | — |
case-22 | pass→pass | 15,186 | 8,601 | -43% | 1 | 1 | 0% | 2,815 | 2,075 | -26% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 19 counted toward the lift figure. The other 3 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +45 percentage points is the difference between those two pass rates over the 19 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.