Get everything flagged on a trace

Returns every detection on one trace, per family (emotion, laziness, hallucination, tool_failure, custom_signal), each with its message anchor (conversation_id, msg_index), plus the Problems the trace is evidence for (id, title, attribution count). No time window. Each family reports detector.enabled (the workspace processing switch, null when there is none) and processed (whether this trace has been through the detector, null when that is not recorded), so an empty family can be told apart from one that did not run.

get/api/v1/data/conversations/{conversationId}/findings

Request

  • Base URL
    https://moda.dev
  • Auth
    API key in header x-api-key

Path parameters

conversationIdstring required

Trace ID (the conversation_id field).

Response

Successful response.

object required

Per-family detections (up to 50 rows each, count is the total), the trace's Problems, and a summary of counts. 404 when the workspace has no such trace.

Changes

No recorded changes to this operation across all 1 revision of this API.