{"slug":"agent-architecture-audit","source_name":"agent-architecture-audit","name":"Agent Architecture Audit","description":"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-first fixes. Essential for developers building agent applications, autonomous loops, or any LLM-powered feature.","version":1,"lift":{"pass_rate_delta_pts":16.67,"pass_rate_pct":16.7,"total_cases":24,"passed_cases":4,"tokens_delta_pct":98.3,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-07-27T16:19:26.479915+00:00"},"skill_score":0.1667,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":16.67,"with_pass_pct":16.7,"without_pass_pct":0,"tokens_delta_pct":98.3,"turns_delta_pct":0,"total_cases":24,"cases_aggregated":23,"verdict":"mixed","never_hurt":true,"completed_at":"2026-07-27T16:19:26.479915+00:00","run_id":"4826b27f-e3c5-4e1f-9c11-5fdf17640ab6","version_number":1,"is_latest_version":true,"gate":null}],"trust":{"skill_safety":"passed","safety_status":"clean","intent_verdict":null,"content_status":"clean","indexable":true},"license":"MIT","install_count":0,"manifest_hash":"72d0f01fcc8f1cf26a0b3f96f40d42cd8a45a29ed2af5a9e15bb3f892251d0f2","raw_url":"https://app.decimal.ai/s/agent-architecture-audit/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/agent-architecture-audit"}