Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Collect Firecrawl debug evidence for support tickets and troubleshooting. Use when encountering persistent issues, preparing support tickets, or collecting diagnostic information for Firecrawl problems. Trigger with phrases like "firecrawl debug", "firecrawl support bundle", "collect firecrawl logs", "firecrawl diagnostic".
.claude/skills/jeremylongshore-firecrawl-debug-bundle/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 1190% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 49% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -21% | 0% |
Produce structured evidence that can reproduce or classify a failure without dumping the repository, environment, request headers, URLs, or response bodies.
Useful evidence includes SDK/runtime version, v2 operation, timestamps, opaque request/job/webhook IDs, HTTP status, documented error/code, queue/job state, pagination state, option names, and content-free metrics. Firecrawl support does not need a broad tarball or the API key.
For authenticated Cloud operations, inject FIRECRAWL_API_KEY from an approved secret manager. REST requests use Authorization: Bearer with the key. Never print, commit, transmit, or place a key in a URL. Keyless access is suitable only where the current documentation explicitly allows it and the workload accepts its limits; production workflows should make identity and team ownership explicit.
Use Read, Glob, and Grep to inspect code, configuration, tests, and evidence. Use Write/Edit only for approved implementation or documentation changes. Do not call Firecrawl, rotate keys, change account settings, scrape a target, or deploy merely because this skill was invoked.
Require approval before collecting from production, including a hostname or request option value, sharing outside the organization, or extending bundle retention.
Return a small manifest and redacted structured evidence files, scan results, omissions, recipient and purpose, expiry, hashes, and cleanup receipt. Do not create a repository or home-directory archive.
Read official Firecrawl evidence before relying on an endpoint, SDK method, plan limit, price, retention option, or self-hosted release.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 31,189 | 18,958 | -39% | 1 | 1 | 0% | 3,877 | 4,328 | +12% | 0 | 0 | — |
case-02 | fail→pass | 9,693 | 22,261 | +130% | 1 | 1 | 0% | 352 | 4,541 | +1190% | 0 | 0 | — |
case-03 | fail→fail | 22,533 | 14,229 | -37% | 1 | 1 | 0% | 4,504 | 3,582 | -20% | 0 | 0 | — |
case-04 | pass→pass | 21,618 | 20,760 | -4% | 1 | 1 | 0% | 4,227 | 4,843 | +15% | 0 | 0 | — |
case-05 | pass→pass | 25,035 | 19,473 | -22% | 1 | 1 | 0% | 3,778 | 4,622 | +22% | 0 | 0 | — |
case-06 | pass→pass | 11,372 | 16,883 | +48% | 1 | 1 | 0% | 2,463 | 4,106 | +67% | 0 | 0 | — |
case-07 | fail→pass | 16,351 | 8,385 | -49% | 1 | 1 | 0% | 2,316 | 3,303 | +43% | 0 | 0 | — |
case-08 | pass→pass | 15,278 | 21,384 | +40% | 1 | 1 | 0% | 2,833 | 4,548 | +61% | 0 | 0 | — |
case-09 | pass→pass | 14,537 | 10,454 | -28% | 1 | 1 | 0% | 1,780 | 3,557 | +100% | 0 | 0 | — |
case-10 | pass→pass | 17,508 | 11,788 | -33% | 1 | 1 | 0% | 3,521 | 4,116 | +17% | 0 | 0 | — |
case-11 | pass→pass | 13,784 | 8,639 | -37% | 1 | 1 | 0% | 1,535 | 2,376 | +55% | 0 | 0 | — |
case-12 | fail→pass | 17,444 | 9,309 | -47% | 1 | 1 | 0% | 2,404 | 3,573 | +49% | 0 | 0 | — |
case-13 | pass→pass | 14,382 | 14,562 | +1% | 1 | 1 | 0% | 1,788 | 3,556 | +99% | 0 | 0 | — |
case-14 | pass→pass | 12,325 | 8,915 | -28% | 1 | 1 | 0% | 2,253 | 3,277 | +45% | 0 | 0 | — |
case-15 | fail→pass | 18,042 | 10,276 | -43% | 1 | 1 | 0% | 3,191 | 2,624 | -18% | 0 | 0 | — |
case-16 | fail→pass | 17,135 | 3,500 | -80% | 1 | 1 | 0% | 2,908 | 2,305 | -21% | 0 | 0 | — |
case-17 | fail→pass | 17,888 | 3,185 | -82% | 1 | 1 | 0% | 3,211 | 2,256 | -30% | 0 | 0 | — |
case-18 | fail→pass | 9,907 | 8,127 | -18% | 1 | 1 | 0% | 1,995 | 2,239 | +12% | 0 | 0 | — |
case-19 | fail→pass | 18,604 | 15,174 | -18% | 1 | 1 | 0% | 2,448 | 3,770 | +54% | 0 | 0 | — |
case-20 | fail→pass | 13,134 | 2,712 | -79% | 1 | 1 | 0% | 1,303 | 2,111 | +62% | 0 | 0 | — |
case-21 | pass→pass | 14,103 | 10,517 | -25% | 1 | 1 | 0% | 2,565 | 3,768 | +47% | 0 | 0 | — |
case-22 | fail→pass | 9,623 | 7,182 | -25% | 1 | 1 | 0% | 988 | 2,081 | +111% | 0 | 0 | — |
case-23 | fail→pass | 18,631 | 3,549 | -81% | 1 | 1 | 0% | 2,520 | 2,279 | -10% | 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 +48 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.