Get a conversation evaluation
One evaluation with its frozen Decision and condition snapshot, input and history, answer, model, usage, the raw provider request and response JSON (from daemons that support it, each capped at 16K characters), and evidence while the daemon still retains the bodies (bounded to 64 KiB); once stripped, detailsExpired is true and only the summary and snapshot remain. Proxied from the serving daemon, never stored or logged, and returned only after the conversation audience check on the namespace the daemon names for the install; before any session there names an audience the bodies need edit access to the consumer agent rather than the organization read baseline. Returns 404 when the evaluation is gone and 503 when the daemon is offline, including while its connection to the install has not yet reported the tenant scope, or must be upgraded, including a reply that names no namespace.
Authorization
bearerAuth In: header
Path Parameters
^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$uuid1 <= length <= 5120 <= value <= 9007199254740991Response Body
application/json
application/json
application/json
curl -X GET "https://example.com/v1/orgs/string/integrations/497f6eca-6276-4993-bfeb-53cbbbba6f08/channels/string/decision-evaluations/0"{ "seq": 0, "at": "string", "messageId": "string", "title": "string", "decisionId": "string", "outcome": "triggered", "reason": "string", "answer": { "type": "boolean", "value": true, "probability": 0 }, "matchedKeys": [ "string" ], "latencyMs": 0, "requestedModel": "string", "actualModel": "string", "usage": { "inputTokens": 0, "outputTokens": 0 }, "detailsExpired": true, "snapshot": { "decisionId": "string", "providerId": "string", "model": "string", "question": { "type": "choice", "instructions": "string", "criteria": { "property1": "string", "property2": "string" } }, "condition": { "type": "choice", "thresholds": { "property1": 0, "property2": 0 } }, "sessionMode": "string" }, "input": { "agent": { "name": "string", "description": "string" }, "currentMessage": { "id": "string", "sender": { "id": "string" }, "text": "string", "quote": { "sender": "string", "text": "string" }, "threadId": "string", "time": "string", "truncated": true }, "history": [ { "id": "string", "sender": { "id": "string" }, "text": "string", "quote": { "sender": "string", "text": "string" }, "threadId": "string", "time": "string", "truncated": true } ], "historyOmitted": 0, "context": { "partial": true, "reasons": [ "string" ], "omittedMessages": 0 } }, "fullAnswer": { "type": "boolean", "value": true, "probability": 0 }, "chain": [ { "stepId": "string", "decisionId": "string", "evaluation": { "status": "answered", "answer": { "type": "boolean", "value": true, "probability": 0 }, "model": "string", "usage": { "inputTokens": 0, "outputTokens": 0 } } } ], "steps": [ { "stepId": "string", "decisionId": "string", "providerId": "string", "model": "string", "question": { "type": "choice", "instructions": "string", "criteria": { "property1": "string", "property2": "string" } }, "condition": { "type": "choice", "thresholds": { "property1": 0, "property2": 0 } }, "rawRequest": { "text": "string", "truncated": true }, "rawResponse": { "text": "string", "truncated": true } } ], "rawRequest": { "text": "string", "truncated": true }, "rawResponse": { "text": "string", "truncated": true }, "evidence": { "snapshotSeq": 0, "suppliedBackground": 0 }}