The Prompting Company

Benchmark report

Task verified

5m 1s elapsed

The saved attempt met the documented success criterion.

Experience score

96/ 100

Agent-Ready
  • SInstall
  • SAuth
  • SExecute
  • ADocs
  • SCurrent
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 SSInstallSAuthSExecuteADocsSCurrent
Agent profile · this run
Judge's note · what slowed the agent most
Hunting for a Clay-managed function routine_id that the Public API exposes no way to list, before pivoting to the query-mode search operation.

Where the time went

5:02 agent time
  • Research0:24
  • Attempt2:50
  • Verify1:48
5m 1sTotal elapsed
1:50Time to first success
0:00Waited on the owner

Judge reason

Saved code (code/enrich_stripe.py, code/enrich_stripe.sh) calls Clay's documented Public API with the clay-api-key header: POST /search/query-mode then POST /search/query-mode/{search_id}/run. Transcript step 78/81 shows the live run printing two HTTP 200s; out/search_create_response.json (200, source_type companies) and out/search_results_response.json (200) hold the live payload with name Stripe, domain stripe.com, industry 'Technology, Information and Internet', size '5,001-10,000', linkedin_url. Field set matches the spec's PublicApiCompanySearchResult exactly (openapi.json), and envelope keys data/has_more/source_type/exhaustion_reason/period_quota match NextQueryModeCompanyResultsResponse, so the shape is documented. The company was identified by domain via clay.include_company_identifiers(("stripe.com")), the exact form the live query reference documents (out/query_reference.md:454). out/run_log.txt printed values match the saved JSON; an independent curl variant (out/curl_variant_output.json) reproduced the same record. No access/bearer token appears in any artifact or log — Clay uses a static key header, kept in /tmp/priv outside home and referenced only via $CLAY_PUBLIC_API_KEY. Nothing is mocked or hard-coded.

Attempt record

  1. 0:59Answered from saved key: Clay API keyno wait
  2. —First success verified

Attempt artifacts saved.