"Techniques for Document Compression and Chunking" matching MCP servers:

An MCP server that exposes documents.js's document conversion, .odb, metadata, and font tooling as MCP tools, enabling agents to convert, inspect, and edit a wide range of document formats over stdio.

Enables document format conversion between Word, Markdown, PDF, HTML, and plain text, supporting batch operations and format validation via the AI MCP protocol.

MCP server for managing product documentation as markdown files. Supports creating, reading, updating, deleting, and searching docs with optional semantic search.

A local document evidence layer for MCP clients that ingests documents (PDF, Office, images) with optional OCR, indexes them in SQLite FTS, and provides retrieval tools with source coordinates.

Provides AI agents with multi-format document indexing, hybrid dense and sparse search with reranking, and relational SQL querying over extracted tables.

Enables clients to inspect local PDFs by searching text, rendering pages, and cropping selected regions at high resolution directly from the source document. It runs locally over stdio without requiring models, embeddings, GPUs, or API keys, and does not modify source PDFs or send network requests.

A local AI document assistant MCP server that enables listing, reading, and editing documents via tools, resources, and prompts, allowing LLMs to manage document workflows through natural language.

A multi-format MCP server that enables reading and generating Office, PDF, text, EPUB, and presentation documents. It provides unified tools for document processing and creation through AI assistants.

Enables AI agents to generate professional Word and PDF documents with support for Markdown, syntax highlighting, and smart pagination. It features automatic JSON detection and responsive A4 formatting for creating high-quality technical reports and manuals.

Enables AI agents to search, deep-read, and build knowledge bases from Markdown, PDF, DOCX, and PPTX documents via MCP tools for retrieval, document navigation, and ingestion.

Converts PDFs, Office files, spreadsheets, emails, audio, and more to Markdown locally, enabling AI assistants to read and process them without cloud upload.

Enables MCP clients to extract text from PDF, image, Word, PowerPoint, and Excel files using MinerU, either on the local machine or a self-hosted private-network API. Parsed results are saved locally as Markdown, structured JSON, and extracted images, which can then be read back in segments.

Enables AI agents to efficiently manage and update component documentation with precise partial updates, saving up to 75% tokens compared to full rewrites.

Local document intelligence for AI agents — extract text, detect tables, read metadata, analyze structure, search keywords, and detect language from PDF and DOCX files. No cloud API required, no API key needed.

Enables real-time indexing and semantic search of local documents (PDF, Word, text, Markdown, RTF) using vector embeddings and local LLMs. Monitors folders for changes and provides natural language search capabilities through Claude Desktop integration.

A no-server MCP toolkit that organizes project wikis from Feishu documents, messages, and more, enabling AI clients to answer questions with cited evidence to prevent hallucinations.

Enables NPC agents to read, write, search, and manage a shared corporate document store, and to turn Mission Scripts into structured report outlines and scaffolds aligned to strategic intent.

Universal MCP server for extracting text from various document formats including PDF, Excel, Word, CSV, and more, with support for streaming, limits, and markdown conversion.

Enables reading and searching documents in allowlisted folders via LiteLLM and MCP tools, using COM for Office files to support IRM-protected content.

Enables AI agents to compress retrieved web documents against a query, pruning irrelevant noise so only the most relevant context reaches the model. It runs as a zero-dependency Python MCP server that accepts a query payload and returns structured, scored output.