mirror of
https://github.com/priyanshujain/messageboardbench.git
synced 2026-10-02 11:07:07 +00:00
620 B
620 B
messageboardbench
Research harness for studying whether coding agents publish, receive, and adopt evaluator-gaming methods across shared scratchpads and message boards.
The current study runs three OpenRouter models on the same 20 ImpossibleBench SWE tasks under isolated, scratchpad, board, feedback, internet, and token-awareness conditions.
See the current experiment, scenario series, experiment index, and setup guide.
Current raw runs live in logs/. Historical local experiments and logs stay in ignored archive/.