Benchmark report

Task verified

3m 56s elapsed

The saved attempt met the documented success criterion.

Experience score

95/ 100

Agent-Ready
  • SInstall
  • SAuth
  • SExecute
  • ADocs
  • ACurrent
Experience measures friction in the tested task. The verdict says whether it succeeded.
Agent profile for this run: Install S, Auth S, Execute S, Docs A, Current ASInstallSAuthSExecuteADocsACurrent
Agent profile · this run
Judge's note · what slowed the agent most
Doc page response-field names (industry, companySize, founded) did not match the live payload (industries, staff, foundedYear), forcing a printer rewrite and a second call.

Where the time went

3:57 agent time
  • Starting0:01
  • Research0:26
  • Attempt2:08
  • Verify1:22
3m 56sTotal elapsed
1:02Time to first success
0:00Waited on the owner

Judge reason

Saved code /workspace/home/enrich_stripe.py POSTs to the documented operation https://api.richapi.ai/api/v1/enrich_company with an x-api-key header read from a private benchmark file, body {"url":"https://www.linkedin.com/company/stripe/"} — the LinkedIn identifier the fetched doc page requires (it states domains and stripe.com-style lookups are not accepted and no domain endpoint exists). Both saved responses show http_status 200 with live server headers (Server: uvicorn, X-Correlation-Id req-51f9165b..., X-Credits-Charged 1) and a payload identifying name "Stripe", website "https://stripe.com", industries ["Computer Software"], staff {total:16694, range 5001-10000}, HQ South San Francisco. run_output.txt prints exactly those values and matches the saved run2 body. Two independent live calls differ only in followers (1806221 vs 1806222), request_id and credit balance — inconsistent with any mock or hard-coded payload. Secret scan of all artifacts found no api key, access token or bearer token; only the placeholder "x-api-key": "[not saved]" is stored.

Attempt record

  1. 0:53Answered from saved key: RichAPI API keyno wait
  2. —First success verified

Attempt artifacts saved.