"How to work with local files" matching MCP servers:

Two self-hosted MCP servers: manage a local model machine (Ollama pull/switch, LoRA training) and bridge to local Ollama/vLLM for pure language processing tasks (writing, summarizing, classifying, extraction) without giving the calling agent tools or file access.

Delegates mechanical text-to-text tasks to a local LLM to save Claude subscription quota, reading files server-side so large content never enters Claude's context.

A local-first MCP server that lets agents search and summarize a user's own WeChat history, with stable pagination, bulk chat workflows, unread/event queries, and gated enrichment tools.

Adds a single MCP tool that lets Codex request local, read-only draft transformations of user-provided text, using a local Ollama model for translation, summarization, rewriting, extraction, and classification.

MCP server that lets Claude Code delegate mechanical tasks to a local LLM for summarization, classification, extraction, and drafting.

MCP server that delegates mechanical tasks like summarization, classification, extraction, and drafting to a local Llama.cpp LLM, serving as a cost-optimization layer while Claude handles reasoning and quality control.

Exposes locally-built AI tools to Claude, enabling article summarization, promo generation, and semantic search over an Obsidian vault.

A Model Context Protocol server that allows AI assistants to discover, load, and process local documents on Windows systems, with support for multiple file formats and OCR capabilities for scanned PDFs.

Integrates local language models (like Qwen3-8B) with MCP clients, providing tools for chat, code analysis, text generation, translation, and content summarization using your own hardware.

Enables local analysis of unstructured documents (PDF, DOCX, PPTX, SVG, PNG) by extracting text and structure with citation anchors, and verifies summaries against source material before a human approves saving a report.

Self-hosted digests of your Telegram channels over MCP: AI summaries, the last digest and raw per-channel messages. Works with OpenAI, Anthropic or local Ollama.

Lets any AI agent score and simplify its own text before it reaches a human, using Flesch readability metrics and plain-language rewrites entirely on the local machine.

An MCP server that provides local, private, synchronous access to Ollama models for prompt answering, text classification, and model listing, without any file access or command execution.

A FastMCP server that enables AI assistants to extract structured information from unstructured text using Google's langextract library through a secure, optimized Model Context Protocol interface.

Enables LLMs to interact with Fathom.video API for managing meeting recordings, retrieving transcripts and AI-generated summaries, searching meeting content, and accessing analytics and team data.

Integrates Vale prose linting into AI coding assistants, enabling users to check text files for style and grammar issues using Vale's powerful linting engine. Provides automated style feedback with smart configuration discovery and rich formatted results.

Enables natural language interaction with local .docx files, allowing users to find, read, search, and summarize Word documents using friendly names and location hints.

Provides MCP-compatible AI clients with offline text analysis and rewriting tools, including statistics, extractive summaries, keywords, readability scores, case conversion, entity extraction, and diffing, all running locally without API keys or network calls.

Enables AI agents and MCP clients to browse Roomtone meeting notes, add YouTube or pasted-text sources, generate summaries, ask questions answered from transcripts, and retrieve raw transcript text. Recording itself stays in the Roomtone app, so the server only reads and adds to existing meetings.

MCP server for HumanPen that lets AI agents work on real documents (.docx, .pptx, .pdf) — humanizing content to lower AI-detection scores, converting citations, condensing, and translating while preserving formatting, tables, images, and citations.

Enables production-grade web research through MCP, combining local SearXNG search with hosted fallback providers, fetch and crawl tools, summarization, and budget guardrails.

Enables reading unstructured meeting notes, generating meeting minutes drafts, validating them against source text, and saving approved versions through a local MCP server.

Enables context compression for Claude Code with Thai language support, reducing token usage by 50-88% while preserving full recoverability of original content.

Turns plain text or Markdown, web pages, and local PDFs into editable Markdown outlines that can be rendered as standalone, offline-capable interactive HTML mind maps. Works with any MCP client with no API key required, since the connected model writes the final outline from a draft and cleaned source text.

Unofficial MCP server for working with Kagi without API access (you'll need to be a customer, tho). Searches and summarizes. Uses Kagi session token for easy authentication.

Analyzes unstructured documents in a local folder, extracting structure and key terms, and supports generating summaries via a host LLM with validated, approval-based report saving.

MCP Long Context Reader is a Python-based toolkit designed to overcome the context window limitations and high costs associated with Large Language Models (LLMs) processing extensive documents. It provides a FastMCP server with multiple, powerful strategies for an LLM agent to 'read' and query long documents without needing to load the entire text into its context window.

Enables any LLM to process arbitrarily long contexts through recursive decomposition, without requiring external LLM APIs.

Enables users to extract and review Claude's plan reasoning in Claude Code by reading plan files and providing a structured template for decisions, tradeoffs, and assumptions.

Provides accurate meeting transcription with speaker diarization and multilingual support, allowing users to submit audio URLs, poll transcription status, get transcripts, and summarize via MCP tools in their IDE.