Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use the `gh` CLI for issues, pull requests, Actions runs, and GitHub API queries.
.claude/skills/sickn33-github/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 34% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -41% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 37% | 0% |
Use the gh CLI to interact with GitHub. Always specify --repo owner/repo when not in a git directory, or use URLs directly.
gh issue, gh pr, gh run, or gh api from the command line.Check CI status on a PR:
bashgh pr checks 55 --repo owner/repo
List recent workflow runs:
bashgh run list --repo owner/repo --limit 10
View a run and see which steps failed:
bashgh run view <run-id> --repo owner/repo
View logs for failed steps only:
bashgh run view <run-id> --repo owner/repo --log-failed
Follow this sequence to investigate a failing CI run:
bash gh pr checks 55 --repo owner/repo
bash gh run list --repo owner/repo --limit 10
bash gh run view <run-id> --repo owner/repo
bash gh run view <run-id> --repo owner/repo --log-failed
The gh api command is useful for accessing data not available through other subcommands.
Get PR with specific fields:
bashgh api repos/owner/repo/pulls/55 --jq '.title, .state, .user.login'
Most commands support --json for structured output. You can use --jq to filter:
bashgh issue list --repo owner/repo --json number,title --jq '.[] | "\(.number): \(.title)"'
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 6,099 | 4,676 | -23% | 1 | 1 | 0% | 1,001 | 693 | -31% | 0 | 0 | — |
case-02 | fail→pass | 5,172 | 4,636 | -10% | 1 | 1 | 0% | 990 | 1,325 | +34% | 0 | 0 | — |
case-03 | pass→pass | 4,115 | 2,805 | -32% | 1 | 1 | 0% | 640 | 1,035 | +62% | 0 | 0 | — |
case-04 | fail→pass | 9,843 | 3,039 | -69% | 1 | 1 | 0% | 1,668 | 980 | -41% | 0 | 0 | — |
case-05 | pass→pass | 3,657 | 3,255 | -11% | 1 | 1 | 0% | 575 | 908 | +58% | 0 | 0 | — |
case-06 | pass→pass | 4,007 | 2,276 | -43% | 1 | 1 | 0% | 560 | 926 | +65% | 0 | 0 | — |
case-07 | fail→pass | 5,961 | 3,286 | -45% | 1 | 1 | 0% | 1,136 | 1,185 | +4% | 0 | 0 | — |
case-08 | fail→pass | 5,928 | 3,087 | -48% | 1 | 1 | 0% | 1,034 | 1,151 | +11% | 0 | 0 | — |
case-09 | fail→pass | 4,072 | 2,147 | -47% | 1 | 1 | 0% | 643 | 882 | +37% | 0 | 0 | — |
case-10 | pass→pass | 5,623 | 3,060 | -46% | 1 | 1 | 0% | 948 | 999 | +5% | 0 | 0 | — |
case-11 | pass→pass | 7,094 | 3,533 | -50% | 1 | 1 | 0% | 1,211 | 1,202 | -1% | 0 | 0 | — |
case-12 | fail→pass | 3,831 | 2,155 | -44% | 1 | 1 | 0% | 627 | 910 | +45% | 0 | 0 | — |
case-13 | pass→pass | 3,266 | 6,779 | +108% | 1 | 1 | 0% | 504 | 1,269 | +152% | 0 | 0 | — |
case-14 | pass→pass | 4,956 | 2,337 | -53% | 1 | 1 | 0% | 808 | 900 | +11% | 0 | 0 | — |
case-15 | pass→pass | 12,005 | 3,348 | -72% | 1 | 1 | 0% | 1,930 | 1,100 | -43% | 0 | 0 | — |
case-16 | pass→pass | 3,474 | 2,423 | -30% | 1 | 1 | 0% | 556 | 875 | +57% | 0 | 0 | — |
case-17 | fail→pass | 5,175 | 2,767 | -47% | 1 | 1 | 0% | 788 | 973 | +23% | 0 | 0 | — |
case-18 | fail→pass | 5,106 | 2,064 | -60% | 1 | 1 | 0% | 851 | 861 | +1% | 0 | 0 | — |
case-19 | pass→pass | 3,010 | 2,004 | -33% | 1 | 1 | 0% | 479 | 824 | +72% | 0 | 0 | — |
case-20 | pass→pass | 10,815 | 7,065 | -35% | 1 | 1 | 0% | 1,686 | 1,548 | -8% | 0 | 0 | — |
case-21 | pass→pass | 4,995 | 4,460 | -11% | 1 | 1 | 0% | 940 | 1,285 | +37% | 0 | 0 | — |
case-22 | pass→pass | 13,246 | 10,159 | -23% | 1 | 1 | 0% | 2,436 | 2,313 | -5% | 0 | 0 | — |
case-23 | pass→pass | 7,678 | 3,965 | -48% | 1 | 1 | 0% | 1,222 | 1,198 | -2% | 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. 23 cases were attempted. The headline lift of +35 percentage points is the difference between those two pass rates over the 23 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.