Start the ScreenJSON REST API server

Start it

screenjson serve --host 0.0.0.0 --port 8080

Optional flags:

Flag	Description
--workers	Worker pool size. Defaults to CPU cores.
--gotenberg	Gotenberg URL for HTML→PDF.
--tika	Apache Tika URL for content extraction.
--llm	OpenAI-compatible LLM endpoint URL.

Endpoints

Method	Path	Purpose
GET	/	Queue status / metrics
GET	/health	Liveness
GET	/formats	Supported formats
POST	/convert	Screenplay file → ScreenJSON
POST	/export	ScreenJSON → screenplay file
POST	/validate	ScreenJSON → { valid, errors[] }

Convert via HTTP

Multipart (simplest for browsers):

curl -X POST http://localhost:8080/convert \
  -F "[email protected]" \
  -F "format=fdx" \
  -F "lang=en"

Octet-stream (cleanest for bots):

curl -X POST http://localhost:8080/convert \
  -H "Content-Type: application/octet-stream" \
  -H "X-Input-Format: fdx" \
  --data-binary @screenplay.fdx

Export via HTTP

curl -X POST http://localhost:8080/export \
  -H "Content-Type: application/json" \
  -H "X-Output-Format: fountain" \
  -d @screenplay.json \
  -o screenplay.fountain

Validate via HTTP

curl -X POST http://localhost:8080/validate \
  -H "Content-Type: application/json" \
  -d @screenplay.json

Returns:

{ "valid": true, "errors": [] }

In Docker

docker run --rm -p 8080:8080 screenjson/cli serve --port 8080

Next

• How-to: Run screenjson-cli in Docker
• How-to: Validate a ScreenJSON document