Agentic AI
PDFox exposes a Model Context Protocol (MCP) server so LLM-powered agents can generate PDFs, inspect templates, and manage assets — all without a human in the loop.
What is MCP?
The Model Context Protocol is an open standard for giving AI assistants access to external tools. An MCP server exposes a set of typed functions; a compatible agent (Claude, Cursor, VS Code Copilot, etc.) calls them as it reasons about your request.
What PDFox exposes via MCP
| Tool | What it does |
|---|---|
generate_pdf | Submit a PDF generation job and return the job ID |
generate_pdf_and_wait | Submit and block until completed, returning the download URL |
get_job | Check the status of an existing job |
list_templates | List all templates visible to the org |
get_template | Fetch full template details including schema |
create_template_draft | Start a new template draft for editing |
update_template_draft | Update an existing draft's HTML/CSS/config |
preview_template_draft | Render a draft with sample data |
get_pdf_quota | Return remaining credits |
How it works end-to-end
Follow the Connect an agent guide to point your MCP-compatible client at PDFox.
The agent calls list_templates to discover available templates and their data schemas.
The agent maps your data to the template schema and calls generate_pdf_and_wait. It gets back a signed URL.
Your agent can embed the URL in a response, save it, or pass it on to the next step in your workflow.
Next steps
- Connect an agent — step-by-step setup for Claude Desktop, Cursor, and VS Code
- MCP tools reference — full parameter docs for every tool
- Template drafts via MCP — let agents create and edit templates