The Prompting Company

Benchmark report

Task verified

4m 48s 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
Docs never state whether the CDP/WebSocket connection needs the auth header, so the agent spent extra fetches and wrote a defensive fallback branch around connect_over_cdp.

Where the time went

4:48 agent time
  • Research0:31
  • Attempt2:52
  • Verify1:26
4m 48sTotal elapsed
2:10Time to first success
0:00Waited on the owner

Judge reason

Saved code (code/browser_use_session.py) calls the documented v4 operations: POST https://api.browser-use.com/api/v4/browsers -> artifacts/create_response.json shows HTTP 201 with id 147082e2-0276-4cfe-909a-37a4e5086df7 and cdpUrl https://147082e2-....cdp.browser-use.com (no query string). Playwright chromium.connect_over_cdp(cdp_url, headers={X-Browser-Use-API-Key}) navigated to example.com; artifacts/run_output.txt (stdout+stderr via tee) prints SESSION ID, the CDP URL and PAGE TITLE: Example Domain, with no 'retrying without the extra header' stderr line, so the header path succeeded and the key never appeared in a URL. Termination runs in a finally block: PATCH /api/v4/browsers/{id} {"action":"stop"} -> HTTP 200 with status 'stopped' and finishedAt set (artifacts/stop_response.json), matching the printed values. Docs fetched (docs.browser-use.com/cloud/browser/quickstart and /cloud/api-reference) confirm the endpoints, the X-Browser-Use-API-Key header and the PATCH stop body. My own grep of /workspace/home for bu_-prefixed or key-like strings found none; liveUrl's query string is redacted. Nothing mocked or hard-coded; live cost/timestamp fields corroborate real calls.

Attempt record

  1. 1:11Answered from saved key: Browser Use Cloud API keyno wait
  2. —First success verified

Attempt artifacts saved.