v51

latestOpenAPI 3.0.0raw.githubusercontent.com2026-08-014102371.0 MB

Trigger an AI audit for a completed AI call.

Evaluates each AI participant in the specified call asynchronously using Gemini. Returns 202 Accepted immediately; poll GET /aiaudits/{id} for completion.

post/aiaudits

Request body

aicall_idstring uuid required

The ID of the AI call to audit.

languagestring

BCP47 language code for audit output (e.g. "en-US", "ko-KR"). Defaults to the call's stt_language or "en-US".

Response

Audit jobs accepted and running in background.

Example response

{
  "result": [
    {
      "id": "550e8400-e29b-41d4-a716-446655440000",
      "customer_id": "7c4d2f3a-1b8e-4f5c-9a6d-3e2f1a0b4c5d",
      "aicall_id": "550e8400-e29b-41d4-a716-446655440000",
      "ai_id": "550e8400-e29b-41d4-a716-446655440000",
      "prompt_history_id": "550e8400-e29b-41d4-a716-446655440000",
      "status": "progressing",
      "overall_score": 4,
      "evaluation": {
        "overall_score": 4,
        "dimensions": {
          "helpfulness": {
            "score": 4,
            "reason": "Provided clear and helpful answers."
          },
          "accuracy": {
            "score": 4,
            "reason": "Responses were factually correct."
          },
          "tone": {
            "score": 5,
            "reason": "Maintained professional and empathetic tone."
          },
          "goal_completion": {
            "score": 4,
            "reason": "Successfully resolved the customer's issue."
          },
          "tool_usage": null
        },
        "summary": "The AI assistant performed well overall."
      },
      "message_ids": [
        "550e8400-e29b-41d4-a716-446655440001",
        "550e8400-e29b-41d4-a716-446655440002"
      ],
      "language": "en-US",
      "error": "LLM API quota exceeded",
      "tm_create": "2026-01-15T09:30:00.000000Z",
      "tm_update": "2026-01-15T09:30:00.000000Z",
      "tm_delete": "2026-01-15T09:30:00.000000Z"
    }
  ]
}