The Prompting Company

Benchmark report

Task verified

6m 8s 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
A 403 proxy block on anchorbrowser.io plus probing GET/DELETE /v1/profiles to learn that profiles are addressed by name, not a returned id.

Where the time went

6:09 agent time
  • Research0:25
  • Attempt3:50
  • Verify1:54
6m 8sTotal elapsed
2:57Time to first success
0:00Waited on the owner

Judge reason

Saved code (code/run.js, code/lib.js) makes real HTTPS calls with the key only in the documented 'anchor-api-key' header (docs/api_profiles_create-profile.md:194), never in a URL; no mocks. Single run (transcript idx 107-110, EXIT=0) matches evidence files byte-for-byte: POST /v1/profiles -> 200 for profile 'bench-tpc-1793746439172' (01), session A id cab2530b-c2b6-4b4f-9122-4a589c57cb0b created with browser.profile{name,persist:true} (03), Playwright connectOverCDP to wss://connect.anchorbrowser.io (apiKey param stripped, header passed) loaded httpbin.org/cookies/set?tpc=benchmark and printed {"cookies":{"tpc":"benchmark"}} (04), DELETE /v1/sessions/{A} -> 200 (05), profile 'source' flipped empty->session (02 vs 06), distinct session B id 1d0f7742-102d-4d83-ba35-93a000887c66 created from the same profile (07), and httpbin.org/cookies in session B returned {"cookies":{"tpc":"benchmark"}} with survived:true (08). Finally block printed DELETE session B -> 200 and DELETE /v1/profiles/{name} -> 200 (09). Grep over evidence/, code/, docs/, result.json found no API key (transcript idx 117, 129); only session UUIDs and rt tokens appear.

Attempt record

  1. 1:28Answered from saved key: Anchor Browser API keyno wait
  2. —First success verified

Attempt artifacts saved.