llm-reliability
Reliability and audit-evidence testing for LLM agents - wrap any agent, assert behavior, measure determinism, check grounding, emit an audit-grade report.
Catch AI code hallucinations—fabricated packages, invented APIs, and contradicted behavior—deterministically, without trusting another model.
Catches safety constraints and task invariants that an LLM agent's context compaction silently drops, then repairs them. Zero dependencies.
