An operating-agent calibration benchmark — measures whether an LLM agent makes correct operating decisions (not whether it can write code).
Latest release v0.11-public-lab · 21 Jun 2026
2 files for Codex, OpenCode and Claude Code: operant AGENTS.md, operant CLAUDE.md — 496 tokens loaded in every session.
AGENTS.md A 491 tok CLAUDE.md A 5 tok These files are saagpatel/operant's own configuration — they tell Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.