Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Collect Bright Data debug evidence for support tickets and troubleshooting. Use when encountering persistent issues, preparing support tickets, or collecting diagnostic information for Bright Data problems. Trigger with phrases like "brightdata debug", "brightdata support bundle", "collect brightdata logs", "brightdata diagnostic".
.claude/skills/jeremylongshore-brightdata-debug-bundle/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 13% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 40% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 46% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 71% | 0% |
Build a deterministic manifest of configuration shape, timestamps, versions, provider codes, and operation identifiers. Default-deny all raw headers and payloads, then include only explicitly allowlisted diagnostic fields.
Read the receipt and Grep candidate logs for secret names, proxy URLs, Bearer values, cookies, query strings, and target content before copying anything.
Write a small metadata file with product, zone alias, client/runtime version, UTC window, operation ID, state, and current provider error fields.
json{"product":"web-scraper-api","zone_alias":"production-redacted","operation_id":"s_redacted","provider_code":"client_10000","raw_payload_included":false}
Use Edit only to remove disallowed material from the private bundle. Use Bash(python:) to scan for configured secret values and forbidden keys; fail closed if any match remains.
Record sorted file hashes, owner, purpose, recipient, and deletion time. Transmit only through the approved private support channel; never attach it to a public issue.
Use Read and Grep for discovery and pre-copy review. Use Write and Edit only inside the exact private bundle path. Use Bash(python:) for local redaction and checksum checks; it must not upload the bundle or call Bright Data.
Include Proxy-Status, x-brd-err-code, a rounded UTC window, and an opaque snapshot ID. Exclude proxy username/password, API key, target URL query, response body, cookies, and collected records.
| Failure | Meaning | Response | |---------|---------|----------| | Secret scan finds a value | Redaction is incomplete | Quarantine the bundle and rebuild from the allowlist | | Raw target content is required to reproduce | Support request exceeds the safe default | Obtain data-owner approval and use a separate controlled transfer | | No operation identifier exists | Evidence cannot be correlated | Reproduce once against an approved test target |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 23,032 | 17,379 | -25% | 1 | 1 | 0% | 4,809 | 5,413 | +13% | 0 | 0 | — |
case-02 | fail→pass | 20,279 | 21,375 | +5% | 1 | 1 | 0% | 4,227 | 5,900 | +40% | 0 | 0 | — |
case-03 | fail→pass | 40,074 | 15,912 | -60% | 1 | 1 | 0% | 4,216 | 5,047 | +20% | 0 | 0 | — |
case-04 | pass→pass | 12,224 | 8,597 | -30% | 1 | 1 | 0% | 2,412 | 3,315 | +37% | 0 | 0 | — |
case-05 | fail→pass | 8,701 | 3,764 | -57% | 1 | 1 | 0% | 1,490 | 2,181 | +46% | 0 | 0 | — |
case-06 | pass→pass | 10,420 | 4,518 | -57% | 1 | 1 | 0% | 1,557 | 2,461 | +58% | 0 | 0 | — |
case-07 | fail→pass | 7,817 | 4,485 | -43% | 1 | 1 | 0% | 1,446 | 2,472 | +71% | 0 | 0 | — |
case-08 | fail→pass | 7,006 | 3,041 | -57% | 1 | 1 | 0% | 1,218 | 2,025 | +66% | 0 | 0 | — |
case-09 | pass→pass | 7,036 | 4,578 | -35% | 1 | 1 | 0% | 1,157 | 2,407 | +108% | 0 | 0 | — |
case-10 | fail→fail | 5,731 | 2,291 | -60% | 1 | 1 | 0% | 1,017 | 1,933 | +90% | 0 | 0 | — |
case-11 | fail→pass | 11,889 | 7,145 | -40% | 1 | 1 | 0% | 2,147 | 2,984 | +39% | 0 | 0 | — |
case-12 | fail→pass | 6,165 | 2,102 | -66% | 1 | 1 | 0% | 949 | 1,813 | +91% | 0 | 0 | — |
case-13 | fail→fail | 7,000 | 7,506 | +7% | 1 | 1 | 0% | 1,220 | 2,844 | +133% | 0 | 0 | — |
case-14 | fail→pass | 10,683 | 5,303 | -50% | 1 | 1 | 0% | 1,963 | 2,483 | +26% | 0 | 0 | — |
case-15 | fail→pass | 14,188 | 9,091 | -36% | 1 | 1 | 0% | 2,416 | 3,164 | +31% | 0 | 0 | — |
case-16 | pass→pass | 12,581 | 5,840 | -54% | 1 | 1 | 0% | 2,191 | 2,450 | +12% | 0 | 0 | — |
case-17 | fail→pass | 11,807 | 4,067 | -66% | 1 | 1 | 0% | 1,763 | 2,347 | +33% | 0 | 0 | — |
case-18 | pass→pass | 7,108 | 2,614 | -63% | 1 | 1 | 0% | 1,183 | 2,030 | +72% | 0 | 0 | — |
case-19 | fail→pass | 19,999 | 6,317 | -68% | 1 | 1 | 0% | 1,313 | 2,836 | +116% | 0 | 0 | — |
case-20 | pass→pass | 8,332 | 7,154 | -14% | 1 | 1 | 0% | 1,532 | 2,961 | +93% | 0 | 0 | — |
case-21 | pass→pass | 16,827 | 19,371 | +15% | 1 | 1 | 0% | 3,215 | 5,204 | +62% | 0 | 0 | — |
case-22 | fail→pass | 7,370 | 5,291 | -28% | 1 | 1 | 0% | 1,353 | 2,554 | +89% | 0 | 0 | — |
case-23 | pass→pass | 5,599 | 4,102 | -27% | 1 | 1 | 0% | 998 | 2,374 | +138% | 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, and 22 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 +57 percentage points is the difference between those two pass rates over the 22 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.