Python CLI and agent plugin that audits whether an AI agent does the job it is supposed to do, using adversarial and operational tests.
Framework where coding agents discover and refine robot sensorimotor skills as reusable code in place of gradient-based training.