Get Artifact

Get one workflow artifact by id.

The artifact kind is derived from the id prefix (extr_… → extraction, clss_… → classification, etc.).

GET /v1/workflows/artifacts/{artifact_id}

Python

from retab import Retab

client = Retab()

artifact = client.workflows.artifacts.get("ceval_abc123")

print(artifact.operation)
print(artifact.id)

200

{
  "operation": "conditional_evaluation",
  "id": "ceval_abc123",
  "run_id": "run_abc123xyz",
  "step_id": "conditional-block-1",
  "selected_handles": ["wrong"],
  "matched_branch_id": "branch_wrong",
  "matched_condition_ids": ["condition_wrong_total"],
  "evaluations": [
    {
      "condition_id": "condition_wrong_total",
      "matched": true,
      "output_handle_id": "wrong",
      "left_value": 1200,
      "operator": "greater_than",
      "right_value": 1000
    }
  ],
  "created_at": "2026-03-12T09:00:04Z"
}

Fetch one dereferenced workflow artifact by artifact_id. The endpoint infers the backing artifact operation from the id prefix, so callers do not need to know which collection stores the record.

The response is the flattened artifact record with operation injected at the top level. Operation-specific fields are preserved.

Python

from retab import Retab

client = Retab()

artifact = client.workflows.artifacts.get("ceval_abc123")

print(artifact.operation)
print(artifact.id)

200

{
  "operation": "conditional_evaluation",
  "id": "ceval_abc123",
  "run_id": "run_abc123xyz",
  "step_id": "conditional-block-1",
  "selected_handles": ["wrong"],
  "matched_branch_id": "branch_wrong",
  "matched_condition_ids": ["condition_wrong_total"],
  "evaluations": [
    {
      "condition_id": "condition_wrong_total",
      "matched": true,
      "output_handle_id": "wrong",
      "left_value": 1200,
      "operator": "greater_than",
      "right_value": 1000
    }
  ],
  "created_at": "2026-03-12T09:00:04Z"
}

Authorizations

Api-Key · string · header · required

Path Parameters

artifact_id · string · required

Response

200

application/json

Successful Response

ExtractionWorkflowArtifact

SplitWorkflowArtifact

ClassificationWorkflowArtifact

ParseWorkflowArtifact

EditWorkflowArtifact

PartitionWorkflowArtifact

ConditionalEvaluationWorkflowArtifact

ReviewEvaluationWorkflowArtifact

WhileLoopTerminationWorkflowArtifact

ApiCallInvocationWorkflowArtifact

FunctionInvocationWorkflowArtifact

An extraction produced by a workflow run, tagged with its artifact operation and creation time.

id · string · required

Unique identifier of the extraction

file · FileRef · object · required

Information about the extracted file

Show child attributes

model · string · required

Model used for the extraction

json_schema · Json Schema · object · required

JSON schema used for the extraction

output · Output · object · required

The extracted structured data

n_consensus · integer · default: 1

Number of consensus votes used

image_resolution_dpi · integer · default: 192

DPI used to render document images

instructions · string | null

Free-form instructions supplied with the extraction request.

status · enum<string> · default: pending

Lifecycle status. The synchronous path returns 'completed'. Background runs progress pending -> queued -> in_progress -> completed | failed | cancelled.

Available options: pending, queued, in_progress, completed, failed, cancelled

error · PrimitiveError · object

Error details when a background run fails; null otherwise. Always present so consumers can read it without an existence check.

Show child attributes

consensus · ExtractionConsensus · object

Consensus metadata for multi-vote extraction runs

Show child attributes

metadata · Metadata · object

Show child attributes

usage · RetabUsage · object

Usage information for the extraction

Show child attributes

created_at · string<date-time>

Timestamp when this artifact was created.

operation · string · default: extraction

The operation that produced this artifact

Allowed value: "extraction"