chimera sandbox-bench

State-based bench: grade the final workspace state + count harmful side effects. Unlike the text benches, this measures what the agent DID (files it changed), and flags mutations outside each task's allowed set. Uses real models + file tools.

Options

--workspace, -wstrdefault: '.sandbox-bench'
Dir to run sandboxed tasks in.
--model, -mstr
Override the model slug.
--max-stepsintdefault: 8
Max tool-calling steps per task.