pdftoolsag/smart-extract-worker repository overview

⁠Smart Extract — Worker

The Smart Extract extraction worker: a service that pulls structured data (fields and line items, with confidence scores and source bounding boxes) out of PDF documents using a vision language model. Exposes an HTTP API with an OpenAPI spec, Prometheus metrics, OpenTelemetry traces, and SSE streaming.

Private Early Access image. Running it requires a valid Smart Extract license key from PDF Tools. This image is distributed for evaluation under the Early Access Program.

• Get started (one-click local stack): https://github.com/pdf-tools/smart-extract-samples⁠
• Product: https://www.pdf-tools.com⁠
• License keys and access: contact PDF Tools sales.

Pair with pdftoolsag/smart-extract-web for the UI.