Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Configure Bright Data CI/CD integration with GitHub Actions and testing. Use when setting up automated testing, configuring CI pipelines, or integrating Bright Data tests into your build process. Trigger with phrases like "brightdata CI", "brightdata GitHub Actions", "brightdata automated tests", "CI brightdata".
.claude/skills/jeremylongshore-brightdata-ci-integration/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 19% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 55% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 183% | 0% |
Keep pull-request verification credential-free and deterministic. Test the Bright Data adapter against fixtures by default, then place any necessary live probe in an independent protected job with a dedicated credential, approved target, and one-operation budget.
Read the workflow and Grep for secrets, direct provider calls, artifact uploads, and execution from untrusted forks. Map every test to either the offline contract lane or the protected live lane.
Write or Edit the workflow so the required job runs npm ci and npm run test:brightdata:contract with BRIGHTDATA_MODE=fixture. Cover success, authentication denial, policy denial, 429, provider 5xx, malformed output, and redaction without network access.
Make the live job independent of fork and pull-request execution. Require a protected environment, maintainer authorization, an approved public target, a dedicated least-privilege secret, a single-operation ceiling, a timeout, and redacted logs.
Use Bash(npm:) to run contract tests with missing and sentinel credentials. Confirm the offline job remains green without secrets and that failure artifacts contain neither credentials nor collected payloads.
Use Read and Grep to inspect workflows, adapters, and secret references. Use Write and Edit for the workflow, fixtures, tests, and runbook. Use Bash(npm:) only for dependency installation and repository test commands; do not use it to make live Bright Data requests.
A forked pull request runs synthetic proxy and snapshot fixtures only. A maintainer-triggered release workflow may perform one approved provider operation through a protected environment and records only status, latency, and redacted error class.
| Failure | Meaning | Response | |---------|---------|----------| | A fork can access a provider secret | Trust boundary is broken | Disable the live job and remove the secret exposure | | Offline tests require the network | Contract lane is nondeterministic | Replace provider traffic with sanitized fixtures | | A live artifact contains raw output | Data boundary is broken | Delete the artifact and retain only redacted metrics |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→pass | 18,151 | 13,432 | -26% | 1 | 1 | 0% | 3,899 | 4,463 | +14% | 0 | 0 | — |
case-02 | fail→pass | 16,088 | 12,959 | -19% | 1 | 1 | 0% | 3,483 | 4,128 | +19% | 0 | 0 | — |
case-01 | fail→pass | 12,656 | 11,968 | -5% | 1 | 1 | 0% | 2,590 | 4,012 | +55% | 0 | 0 | — |
case-04 | pass→pass | 12,001 | 4,125 | -66% | 1 | 1 | 0% | 2,257 | 2,192 | -3% | 0 | 0 | — |
case-05 | fail→pass | 9,850 | 3,323 | -66% | 1 | 1 | 0% | 1,835 | 2,009 | +9% | 0 | 0 | — |
case-06 | fail→pass | 13,053 | 2,619 | -80% | 1 | 1 | 0% | 663 | 1,879 | +183% | 0 | 0 | — |
case-07 | fail→fail | 9,839 | 2,993 | -70% | 1 | 1 | 0% | 1,837 | 1,880 | +2% | 0 | 0 | — |
case-08 | fail→pass | 14,236 | 3,088 | -78% | 1 | 1 | 0% | 2,667 | 1,785 | -33% | 0 | 0 | — |
case-09 | fail→fail | 11,075 | 9,678 | -13% | 1 | 1 | 0% | 2,149 | 3,559 | +66% | 0 | 0 | — |
case-10 | pass→pass | 11,212 | 8,230 | -27% | 1 | 1 | 0% | 2,075 | 3,049 | +47% | 0 | 0 | — |
case-11 | pass→pass | 13,137 | 6,082 | -54% | 1 | 1 | 0% | 2,059 | 2,466 | +20% | 0 | 0 | — |
case-12 | fail→fail | 13,993 | 9,065 | -35% | 1 | 1 | 0% | 2,612 | 3,077 | +18% | 0 | 0 | — |
case-13 | fail→pass | 12,790 | 4,047 | -68% | 1 | 1 | 0% | 2,128 | 2,122 | -0% | 0 | 0 | — |
case-14 | pass→pass | 8,451 | 3,752 | -56% | 1 | 1 | 0% | 1,411 | 1,987 | +41% | 0 | 0 | — |
case-15 | pass→pass | 8,815 | 2,920 | -67% | 1 | 1 | 0% | 1,629 | 1,750 | +7% | 0 | 0 | — |
case-16 | pass→pass | 9,439 | 7,545 | -20% | 1 | 1 | 0% | 1,743 | 2,852 | +64% | 0 | 0 | — |
case-17 | pass→fail | 10,704 | 5,820 | -46% | 1 | 1 | 0% | 2,076 | 2,493 | +20% | 0 | 0 | — |
case-18 | pass→pass | 12,317 | 5,879 | -52% | 1 | 1 | 0% | 2,495 | 2,620 | +5% | 0 | 0 | — |
case-19 | pass→pass | 6,314 | 2,342 | -63% | 1 | 1 | 0% | 1,180 | 1,835 | +56% | 0 | 0 | — |
case-20 | pass→pass | 21,801 | 36,780 | +69% | 1 | 1 | 0% | 3,804 | 5,823 | +53% | 0 | 0 | — |
case-21 | pass→pass | 17,033 | 14,822 | -13% | 1 | 1 | 0% | 3,357 | 4,721 | +41% | 0 | 0 | — |
case-22 | pass→pass | 16,808 | 16,475 | -2% | 1 | 1 | 0% | 3,218 | 4,657 | +45% | 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 21 counted toward the lift figure. The other 1 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 +27 percentage points is the difference between those two pass rates over the 21 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
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.