PaperPilot

// Overview

PaperPilot is an open-source PDF and document toolkit written in Rust. A single engine powers five surfaces: an MCP server for AI agents, a command-line tool for batch jobs, a WebAssembly build that runs in the browser tab, a Tauri and Svelte desktop app with a viewer, markup and diff, and a local REST API on port 7823. The project's promise is that files stay on your machine — the browser version processes documents in the tab without uploading them.

The toolset covers the everyday PDF jobs people usually hand to an online converter: merging, splitting, rotating, reordering and extracting pages, booklet and N-up layouts, lossy compression with stream deduplication, watermarks, signature validation, a PII scan and deep sanitising. For AI pipelines it extracts structured text and Markdown tables, runs OCR with PaddleOCR on the pure-Rust `rten` runtime, and exposes Tika-style endpoints so tools such as Open WebUI can use it as a document extraction server.

The MCP server is the agent-facing part: it runs as a local stdio server and can print ready-made client configuration for Codex, Claude Code, Kimi Code, Pi, Claude Desktop and Cursor, so an assistant can act on a request like "merge these two PDFs and compress the result" without sending the files to a cloud service. The repository is the Apache-2.0 Community Edition; an Enterprise Edition with SSO, RBAC, audit logs and DLP is developed separately under a commercial licence.

// What it does

• One Rust engine shared by the MCP server, CLI, browser WASM build, Tauri desktop app and local REST API
• Page organisation and layout: merge, split, rotate, delete, reorder, extract, booklet imposition, N-up grids and poster tiling
• Compression with lossy image settings, stream deduplication and CMYK safety guards
• Structured text and Markdown-table extraction plus pure-Rust OCR (PaddleOCR on `rten`)
• Tika-compatible endpoints (`/tika/text`, `/tika/json/md`, `/rmeta/text`) for Open WebUI and other RAG ingestion
• Security tools: watermarking, signature validation and removal, PII scanning and deep sanitising
• `paperpilot-mcp --client-config` prints setup for Codex, Claude Code, Kimi Code, Pi, Claude Desktop and Cursor

// Getting started

Pre-built `paperpilot-mcp` binaries are on the GitHub Releases page, and the browser version runs at usepaperpilot.com with no install. To build everything from source you need Rust 1.75+ (plus Node.js and pnpm for the desktop app).

git clone https://github.com/KrushnaVardhanReddy/PaperPilot.git
cd PaperPilot
cargo build --release -p paperpilot-cli -p paperpilot-mcp -p paperpilot-gateway

paperpilot-mcp --client-config codex

paperpilot merge --inputs doc1.pdf doc2.pdf --output merged.pdf
paperpilot compress --input in.pdf --quality 75 --output compressed.pdf
paperpilot rotate --input in.pdf --pages 1,3-5 --angle 90 --output out.pdf

paperpilot serve --port 7823
curl -X PUT http://localhost:7823/tika/text \
  -H "Content-Type: application/pdf" \
  --data-binary @sample.pdf

cd apps/desktop && pnpm install && pnpm tauri dev

Commands and code are distilled from the project's own documentation — always check the official repo for the latest.

// When to use it

• Let Claude Code, Codex or Cursor merge, compress or OCR PDFs through MCP without uploading confidential documents
• Replace online PDF converters with a browser tool that never sends the file to a server
• Feed scanned or text PDFs into a RAG pipeline through the Tika-compatible extraction endpoints
• Script bulk page operations and compression in CI or n8n through the CLI or the local REST API

// How PaperPilot compares

PaperPilot alongside other open-source office documents & spreadsheets tools AI/TLDR tracks, ranked by GitHub stars.