Python CLI and agent plugin that audits whether an AI agent does the job it is supposed to do, using adversarial and operational tests.
Jupyter notebooks and recipes demonstrating code patterns for building with the Claude API, including tool use, RAG, vision, and evaluations.