Document Processing is an MCP server published by ai.filegraph that connects an agent to the FileGraph API for document operations. It exposes capabilities over a streamable HTTP transport. The server lets an agent extract text from documents, manipulate PDF files, and perform optical character recognition. These operations run through the FileGraph API rather than local libraries. Because the tool list was not provided in the registry record, the specific endpoint or method names available to an agent are not documented here. The upstream source code for the integration is available at the repository URL https://github.com/filegraph/docconvert. This capability is an appropriate choice when an agent needs to read content from documents, convert scanned images into machine-readable strings, or adjust existing PDF files as part of an automated workflow. It is not suitable for use cases that require offline processing without an API key, since all functionality routes through the FileGraph API. It also does not expose document-generation capabilities beyond PDF manipulation, and it requires active network connectivity to the FileGraph service.
Document Processing
Extract text from documents, manipulate PDFs, and perform OCR via FileGraph API.
Use cases
- Extract text from uploaded documents for agentic analysis
- Perform OCR on scanned PDFs via the FileGraph API
- Manipulate existing PDF files within an automated workflow
When to use it
- You need an MCP server that performs OCR and PDF manipulation
- Your agent already uses or can reach the FileGraph API
- You prefer to offload document processing to a remote API rather than local tools
When not to use it
- You need a fully offline document processor without external API dependencies
- You require specific tool names or local binaries not provided here
- You need to create new documents from scratch rather than manipulate existing PDFs