Produce documents by API

Applications call DocIbis over HTTPS with an API key. The interactive reference of every call is in the console, at /docs.

Produce one document

POST /v1/render takes a form (multipart/form-data): the template, by its name in the library (templateName) or as a file (template), and the JSON data (data). The answer is the document.

curl -H "Authorization: Bearer $DOCIBIS_API_KEY" \
  -F templateName=exemples/03-facture.docx \
  -F data=@invoice.json \
  -o invoice.pdf https://docs.example.com/v1/render

Useful options, as form fields:

Field	Effect
format=docx	a Word document instead of a PDF
outputName=invoice-0512.pdf	the file name given to the document
pdfa=PDF/A-2b	an archival PDF (PDF/A-1b, 2b or 3b)
accessible=true	a tagged PDF, for screen readers
watermark=COPY	a watermark on every page
passwordProtect=…	a password to open the PDF

The X-Render-Ms header gives the time taken. Errors come back as JSON (application/problem+json) with a readable detail.

Several documents at once

• Batch: POST /v1/render/batch produces many documents in one call and returns a ZIP file.
• Bundles: a bundle groups several templates and fixed PDF files (terms and conditions, for example) into one PDF; POST /v1/bundles/{name}/render.

Coming from Docmosis

DocIbis answers the Docmosis call POST /api/render with the same fields (accessKey, templateName, outputName, data and the PDF options). Point the client to DocIbis, give it a DocIbis API key as accessKey, and import its templates in the library. Compare a few documents before switching: the reference documents of each template do it page by page.

Integrations

The Integrations page gathers what an application or an AI assistant needs: the address of the MCP server, its configuration for Claude, VS Code or Cursor, the keys and the guides.

• AI assistants: the MCP server at /v1/mcp lets an assistant list templates and produce documents with the rights of its API key.
• Monitoring tools: Prometheus metrics at /metrics, protected by METRICS_TOKEN.