The Prompting Company

Benchmark report

Task verified

6m 17s elapsed

The saved attempt met the documented success criterion.

Experience score

100/ 100

Agent-Ready
  • SInstall
  • SAuth
  • SExecute
  • SDocs
  • 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 S, Current SSInstallSAuthSExecuteSDocsSCurrent
Agent profile · this run
Judge's note · what slowed the agent most
The research plan named BTC-USDT, so the first placement hit a 403 policy gate and the agent had to confirm no order existed and re-derive params for BTC-USDC.

Where the time went

6:18 agent time
  • Research0:28
  • Attempt3:39
  • Verify2:10
6m 17sTotal elapsed
2:04Time to first success
0:00Waited on the owner

Judge reason

Transcript shows every exchange call via mcp__okx__* tools on okx-trade-mcp 1.5.0 with demo:false (live). BTC-USD spot returned 51001, so BTC-USDC was the correct fallback pair; tickSz 0.1, lotSz 0.00000001, minSz 0.00001 and best bid 73546.7 were read from market_get_instruments/market_get_orderbook and printed, with an empty pre-existing open-order listing. Three separate spot_place_order calls (transcript lines 77-84) returned sCode 0 and ordIds 4092543234357997569 / 4092543376561680385 / 4092543515443474433 for clOrdIds ctrl0mo1/2/3 + shared suffix 1794469649; prices 58837.3, 58101.8, 57366.4 are exactly the 20/21/22% targets (58837.36, 58101.893, 57366.426) rounded down one tick, and notionals 10.0006, 10.0005, 10.0001 USDC all sit in [10,11). spot_get_orders at 07:48:23 listed all three live with accFillSz 0; one spot_batch_cancel (POST /api/v5/trade/cancel-batch-orders) returned sCode 0 for all three; the follow-up listing returned data:[] and per-order spot_get_order lookups show state canceled, accFillSz 0, fee 0. Bash was used only for decimal arithmetic and evidence files, never against the exchange; no credential strings appear in any saved file or output. An initial BTC-USDT placement (the pair the research plan named) was rejected 403 by the policy gate and a clOrdId lookup returned 51603, proving no order was created, so exactly three orders exist.

Attempt record

  1. —First success verified

Attempt artifacts saved.