Skip to main content

Orchestrator API

The Orchestrator API exposes the higher-level redaction workflow on top of AI Smart Redact: each upload becomes a managed job that automatically runs detection, waits for review, runs redaction, and stores the redacted PDF for download. Authentication, status tracking, request queueing, and access control are built in. For direct, low-level integration with the underlying file, detection, and redaction operations, use the Manager API instead.

Authentication

The Jobs endpoints authenticate with an API key sent in the X-Api-Key header, or a Bearer JWT in the Authorization header. API keys are created and managed by an administrator; refer to the Authentication guide for how to create one.

API keys have the format pdft_ followed by a base64url-encoded random suffix. Each key has a required expiration of up to 12 months and is shown in plain text only once at creation; if the key is lost, a new one must be created. Revocation takes effect on the next request.

Requests without a valid API key or JWT receive a 401 Unauthorized response. The per-endpoint status code tables don’t repeat this.

API versioning

All Orchestrator API endpoints are prefixed with /v1.

Error responses

When a request fails, the response body is an ApiErrorResponse. Per-endpoint pages list the status codes each endpoint returns; the body shape is the same in all cases.

FieldTypeDescription
typestringURI identifier for the error type.
titlestringShort summary of the error.
statusintegerHTTP status code.
traceIdstringCorrelation ID for the request, useful for support and log lookup.
detailstringDescription of the error.
errorsobject of string arraysField-level validation errors, keyed by field name.

Reference pages