Get a prompt run by ID

Retrieves the current state of a prompt run, including status, outputs, token usage, cost, and errors. Poll this endpoint after receiving a webhook notification, or use it to check the status of a run in progress.

GET /api/v1/promptruns/{id}

PromptJuggler PHP SDK

PHP

$run = $pj->getPromptRun('01J...');

200

{
  "id": "3c90c3cc-0d44-4b50-8888-8dd25736052a",
  "status": "pending",
  "createdAt": "2023-11-07T05:31:56Z",
  "emitted": [
    {
      "tool": "<string>",
      "payload": {}
    }
  ],
  "transcript": [
    {
      "content": "<string>",
      "citations": [
        {
          "title": "<string>",
          "url": "<string>"
        }
      ],
      "type": "text"
    }
  ],
  "finishedAt": "2023-11-07T05:31:56Z",
  "output": "<string>",
  "error": "<string>",
  "tokenUsage": {
    "input": 123,
    "inputCached": 123,
    "output": 123,
    "reasoning": 123,
    "total": 123,
    "serviceTier": "auto",
    "inputCacheWrite": 0
  },
  "cost": {
    "success": {
      "input": 123,
      "cachedInput": 123,
      "cacheWrite": 123,
      "output": 123,
      "webSearch": 123,
      "total": 123
    },
    "retries": {
      "input": 123,
      "cachedInput": 123,
      "cacheWrite": 123,
      "output": 123,
      "webSearch": 123,
      "total": 123
    },
    "total": 123
  }
}

Authorizations

Authorization · string · header · required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Path Parameters

id · string<uuid> · required

Prompt run ID

Response

200

application/json

Prompt Run

Prompt run status and result.

id · string<uuid> · required

Prompt run ID.

status · enum<string> · required

Current run status.

Available options: pending, completed, failed

createdAt · string<date-time> · required

Timestamp when the run was created.

emitted · object[] · required

Payloads produced by emit tools during this run, in call order. Empty until the run completes.

Show child attributes

transcript · (TranscriptText · object | TranscriptTool · object | TranscriptData · object)[] · required

The run as a renderable sequence: assistant text, the tools it used, and emit payloads in position. Empty until the run completes. output remains the flat text for callers that only need the answer.

A block of assistant text.

TranscriptText

TranscriptTool

TranscriptData

Show child attributes

finishedAt · string<date-time> | null

Timestamp when the run finished. Null while the run is pending.

output · string | null

LLM output text produced so far; read status for completeness. Null when the run failed, or when the model returned no text — e.g. a turn that was only tool calls or only reasoning.

error · string | null

Error message from the latest failed attempt, kept even once a retry recovers — so a pending or completed run can carry one. Read status for the outcome.

tokenUsage · object | null

Token usage accumulated over successful turns — a run that failed later still reports the earlier ones. Null until the first turn succeeds.

Show child attributes

cost · object | null

Cost breakdown for the run. Null while pending, or when no published rate covers the run.

Show child attributes