Full-stack diagnostic for agent and LLM applications. Audits the 12-layer agent stack for wrapper regression, memory pollution, tool discipline failures, hidden repair loops, and rendering corruption. Produces severity-ranked findings with code-fi... Use when auditing an AGENT APPLICATION's overall design and stack. Do NOT use for: comparing agent tools (→ agent-eval), debugging a single agent run (→ agent-introspection-debugging), or rating completed task output (→ agent-self-evaluation).
