{
  "term": {
    "id": "lex:agent-escalation-and-correction-loops",
    "slug": "agent-escalation-and-correction-loops",
    "term": "Agent Escalation & Correction Loops",
    "aliases": [
      "Agent Escalation and Correction Loops"
    ],
    "kind": "canon",
    "type": "concept",
    "definition": "Protocol for detecting agent output failure, triggering escalation to a higher-tier agent or human review, and applying correction logic before the pipeline continues. Enables self-correcting agent chains that don't fail silently. Core quality control layer in all Neuruh SAL deployments.",
    "definition_basis": "verbatim",
    "definition_source": "https://app.notion.com/p/339a981d75b681d68b58d433d1132995",
    "provenance": [
      {
        "source": "notion",
        "ref": "https://app.notion.com/p/339a981d75b681d68b58d433d1132995",
        "captured_at": "2026-07-25"
      },
      {
        "source": "notion",
        "ref": "https://app.notion.com/p/5a77422a5a0e46b7a7d5b7521ba7e42e",
        "captured_at": "2026-10-05"
      }
    ],
    "related": [],
    "status": "published",
    "first_seen": "2026-04-04",
    "source_ids": [
      "ip:CORE-PROTO-ESCLOOP-1.0-BUILT"
    ]
  },
  "url": "https://lexicon.neuruh.com/t/agent-escalation-and-correction-loops",
  "cite": "lex:agent-escalation-and-correction-loops — Neuruh Lexicon v1.0.0",
  "lexicon_version": "1.0.0",
  "content_hash": "0a2bb72426e976507b89a3172a8c5671b6fe136f6e54ce8499ec1d890ac7e851"
}