video1.0<iframe src="https://www.loom.com/embed/d6efd890aede4c56aef11a829c053643" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/os/loom-media-content/loom/dd6b0ea0-1c25-416b-94fd-bc761c0e1b7a/54203b7f-b878-4302-8d7d-6806bcbe12cc/sessions-thumbnails-d6efd890aede4c56aef11a829c053643-87ea51ec151b1cdc.gif/binary?authToken=eyJraWQiOiJtaWNyb3MtZ3JvdXAvdGRwLW9zLWFyY2hldHlwZS9vZXE4Z28waTlyYjRkaHZrIiwiYWxnIjoiUlMyNTYifQ.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.CFQPoYPqPNouDZ4j9V4Jz24q-_btkVWjPTV75flokFFCBTBp8NNaUajIDfVEMvCMpdXxd4jHKd1AWBpyOuFmww85HuiJg5gY9NP1GWg26v8uBxkJWCHmSTIjXTadFxcj1Ms3uqS9NE8-oh2F4DsW7lE-pWE1gjUJk4wMKsDX-o7n6c-3D0sqXt9a9vVyCUF9jwRRQC78VI6eyown2zzsPAUOB_Cw9EeW4GCe7AVVLpZvwK87uGl3SdYNSdDTGa5aMztCpKHjjvKNdHfeCyxXk0240bem4G1T0xfqbH4oK0ZeF9uC4vLcWJ-NKX52XWlewlow6iCLbFEeyVznST9-Tw&Expires=1788800041&Signature=EhN-VdWaxlmUs93nX3rcrWRIoxscqCPuhLb6QOrbAJRdNuVTxCXaGdsLLGHqzeP8PICSuc2O1f-dmujAnfw6g66EINpmVbCIY7u7VcIud1Pq3pICWi2JhUaDVyNs5reXQb68FAQeLon4MASiyPCLmfj-oMIKV9RV8dKF2NNVFrEFWqBUl9aramusfD9K3y33Z2feQnkGIWe4WC%7EVLN52ibZmEermjUnc7NbdDy5%7EohaLJ5iBu%7E1y4xWl0kNpdsA9nR7h2BNzYIP-r2fluA8ua36p1Vp1G23fAuSQaduHfZNeW4qPcrdfiCX5sTtAR2bY-ZJlE5%7EgeVAU%7EcWm1Cu%7EMw__&Key-Pair-Id=KQOSYIR44AIC0253.69This Loom explains the Incident Review Harness by TechTricksters, a six-agent workflow for producing policy-checked, audit-ready incident reports with clear reasoning behind every step. It covers six stages: intake parsing, root cause analysis with ranked confidence (not certainty), policy or SOP checking against fixed knowledge, a critic to flag unsupported high-confidence claims, an escalation card to determine resolution versus escalation, and a report writer that assembles the final plain text audit report. A live demo shows the pipeline running after selecting an API condition, producing a structured report that identifies issues like a flat out of tolerance temperature during 2 PM shifts. The output includes likeliest cause, policy relevance, recommendations, confidence cabinets, and similar incidents with a percentage match.