jztan/pdf-mcp — MCP server that lets Claude Code and other AI agents read large PDFs without hitting context limits.

Description and use

jztan/pdf-mcp is a lightweight backend service that enables Claude Code and other AI agents to ingest and query massive PDF documents without exceeding token limits. It preprocesses files into searchable chunks, combines vector and keyword search, and adds OCR, table and image extraction while caching results in SQLite for fast repeated access. The project is ideal for developers needing AI‑driven document retrieval and analysis in their applications.

Problem

jztan/pdf-mcp helps connect AI assistants to real tools and data through a standard protocol.

Outcome

Medium: useful for prototypes or internal workflows, with dependency and maintenance checks before production.

Workflow

• connect AI agents to tools
• ship Model Context Protocol servers
• standardize integrations

Evidence

MCP server that lets Claude Code and other AI agents read large PDFs without hitting context limits. Chunked reading, hybrid search, OCR, table and image extraction, SQLite cache.

Source