Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Trigger a CI workflow for a package, wait for completion, and report results
.claude/skills/tetherto-ci-validate/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 926% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 40% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 85% | 0% |
Trigger a CI workflow for a package, wait for it to complete, and report results.
The argument $ARGUMENTS is the package short name. See the CI Package Mapping table in .agent/knowledge/ci-validation.md for valid short names and their corresponding package directories.
The workflow name is: On PR Trigger (<short-name>) where <short-name> is the argument.
$ARGUMENTS matches one of the valid package names above. If not, show the list and stop.git push origin HEADRun: gh workflow run "On PR Trigger ($ARGUMENTS)" --repo tetherto/qvac --ref $(git branch --show-current)
If the trigger fails, show the error and stop.
Wait 5 seconds for the run to register, then find the run ID: gh run list --repo tetherto/qvac --workflow "On PR Trigger ($ARGUMENTS)" --branch $(git branch --show-current) --limit 1 --json databaseId,status,conclusion,createdAt
Wait for the run to complete: gh run watch <run-id> --repo tetherto/qvac
After the run completes, get the result: gh run view <run-id> --repo tetherto/qvac
gh run view <run-id> --repo tetherto/qvac --log-failedFor failure analysis, platform details, and troubleshooting: see .agent/knowledge/ci-validation.md.
Important: Only attempt to fix infra/CI failures (environment, config, workflow issues). If the failure is a code logic error (compilation error, test assertion, lint violation, type error), report the failure back to the user with the relevant logs — do not attempt to fix application code.
If running as part of an automated pipeline (not interactive):
.agent/knowledge/ci-validation.md| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 7,304 | 4,428 | -39% | 1 | 1 | 0% | 1,191 | 932 | -22% | 0 | 0 | — |
case-02 | fail→fail | 2,766 | 4,179 | +51% | 1 | 1 | 0% | 351 | 972 | +177% | 0 | 0 | — |
case-03 | fail→fail | 7,014 | 4,728 | -33% | 1 | 1 | 0% | 1,153 | 888 | -23% | 0 | 0 | — |
case-04 | fail→fail | 4,121 | 4,222 | +2% | 1 | 1 | 0% | 719 | 891 | +24% | 0 | 0 | — |
case-20 | pass→pass | 13,533 | 9,579 | -29% | 1 | 1 | 0% | 2,338 | 2,384 | +2% | 0 | 0 | — |
case-05 | fail→fail | 4,134 | 5,310 | +28% | 1 | 1 | 0% | 678 | 959 | +41% | 0 | 0 | — |
case-06 | pass→fail | 5,977 | 5,065 | -15% | 1 | 1 | 0% | 1,024 | 893 | -13% | 0 | 0 | — |
case-07 | fail→pass | 6,134 | 6,312 | +3% | 1 | 1 | 0% | 181 | 1,857 | +926% | 0 | 0 | — |
case-08 | fail→pass | 8,659 | 3,443 | -60% | 1 | 1 | 0% | 1,477 | 1,273 | -14% | 0 | 0 | — |
case-09 | fail→pass | 11,353 | 2,193 | -81% | 1 | 1 | 0% | 1,578 | 1,057 | -33% | 0 | 0 | — |
case-10 | fail→pass | 4,626 | 2,617 | -43% | 1 | 1 | 0% | 788 | 1,105 | +40% | 0 | 0 | — |
case-11 | pass→pass | 10,307 | 2,459 | -76% | 1 | 1 | 0% | 1,763 | 1,099 | -38% | 0 | 0 | — |
case-12 | pass→pass | 5,304 | 3,025 | -43% | 1 | 1 | 0% | 876 | 910 | +4% | 0 | 0 | — |
case-13 | fail→fail | 8,242 | 2,117 | -74% | 1 | 1 | 0% | 1,306 | 1,024 | -22% | 0 | 0 | — |
case-14 | fail→pass | 3,161 | 1,603 | -49% | 1 | 1 | 0% | 509 | 942 | +85% | 0 | 0 | — |
case-15 | fail→pass | 5,584 | 1,931 | -65% | 1 | 1 | 0% | 922 | 959 | +4% | 0 | 0 | — |
case-16 | fail→pass | 15,141 | 4,046 | -73% | 1 | 1 | 0% | 2,328 | 1,473 | -37% | 0 | 0 | — |
case-17 | pass→pass | 8,539 | 4,253 | -50% | 1 | 1 | 0% | 1,511 | 1,443 | -5% | 0 | 0 | — |
case-18 | fail→pass | 5,405 | 1,968 | -64% | 1 | 1 | 0% | 725 | 927 | +28% | 0 | 0 | — |
case-19 | pass→pass | 13,147 | 2,084 | -84% | 1 | 1 | 0% | 2,070 | 985 | -52% | 0 | 0 | — |
case-21 | pass→pass | 6,584 | 3,180 | -52% | 1 | 1 | 0% | 1,139 | 1,241 | +9% | 0 | 0 | — |
case-22 | pass→pass | 10,940 | 5,296 | -52% | 1 | 1 | 0% | 1,818 | 1,581 | -13% | 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 15 counted toward the lift figure. The other 7 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 +32 percentage points is the difference between those two pass rates over the 15 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
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.