MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
Agent-first OCR CLI and MCP server. Extract images, PDFs, and URLs with Gemini, Kimi, Muse, OpenRouter, or compatible APIs.
npm MIT -
Agent-first OCR CLI and MCP server for images, PDFs, URLs, structured data, and resumable extraction with Gemini, Kimi, Muse, OpenRouter, and compatible APIs.
npm MIT ★ 6 -
auto-reader-ocrgithub.com/Samqra/auto-reader-ocrupdated this year
Arabic-first OCR, translation & document extraction API — MCP server for Claude Code & agents. Keyless: first call mints a free trial key.
MIT -
OCR MCP server for ocr-software.com: let Claude, Cursor and other agents read PDFs and scans
pypi MIT -
AxiomVision — Deterministic OS-Level Vision, Sub-Pixel OCR & Accessibility Automation Engine for AI Agents.
MIT ★ 1 -
Fast PDF operations for AI agents: CLI and MCP server (text, tables, OCR, redaction, scanning, forms, signatures, stamps, merge, split)
npm MIT -
my_tiny_agentgithub.com/yanjin101/my_tiny_agentupdated this year
Spec-driven small LLM agent that operates a real desktop end-to-end (OCR + YOLO + Win32), no vision LLM required. 规范驱动的端到端桌面操作 Agent。
MIT -
Pagecraft Tools for AI agents: 32 web data tools (site audits, scrapers, OCR, PDF text) via the Apify MCP server. Pay per result.
-
One MCP connector for 420+ vetted Agent Skills run in EU sandboxes: real Office files, OCR and tested business methods from any assistant.
MIT -
本地视觉理解 MCP 服务，为 AI agent 提供识图能力：通过 vision / ocr / list_models / providers 四个 MCP 工具描述截图、界面、图表与照片。
MIT ★ 5 -
mcp-serial-hid-kvmgithub.com/sunasaji/mcp-serial-hid-kvmupdated this year
MCP server that gives AI agents (Claude, etc.) full keyboard, mouse, and screen control of a physical PC via serial-hid-kvm. Includes OCR-based screen reading.
MIT ★ 6 -
45 micro-priced utilities for AI agents (x402 USDC on Base). Search, screenshots, OCR, PDF invoices, whois, geoip, prices. MCP server + LangChain adapter included.
-
Local MCP server: turn screen recordings into transcript, exact frames, OCR and wall-clock evidence for coding agents.
pypi MIT -
Computer use for every Windows app window, for any MCP agent: GPU-rendered, covered or in the background. UI Automation tree, capture fallback chain, OCR, background input.
Apache-2.0 -
Open-source MCP vision server that gives text-only LLMs and AI agents image understanding, OCR, visual analysis, UI inspection, and multimodal capabilities.
MIT ★ 1 -
Visual bridge for text-only models and visual enhancement toolkit for multimodal AI agents — OCR, zoom, PDF/Office rendering, web vision and interaction.
★ 20 -
paddleocr_mcpgithub.com/YuanJinke/paddleocr_mcpupdated this year
基于 PaddleOCR 的 MCP 服务器，完全免费，为 AI Agent 提供本地图片文字识别能力。无需 API Key，支持中英文，完全离线运行。Free PaddleOCR-based MCP Server: Offline Chinese & English OCR for AI Agents, no API Key needed.
MIT ★ 3 -
win32-mcp-servergithub.com/randynorthrup/win32-mcp-serverupdated this year
Enterprise-grade Windows automation MCP server for AI agents — 53 tools including click_text, wait_for_text, structured OCR, UI Automation, and batch operations
pypi MIT -
Control any Windows window or browser from an LLM agent: UI Automation + vision/OCR + real input + CDP, packaged as a DSH plugin / MCP server with 37 tools.
MIT -
Agentic RAG over one PDF or a whole folder, with page-level evidence to check every answer: hybrid search, selective page reads, tables, OCR, charts, and CJK layouts.
pypi MIT ★ 171 -
Zotero AI plugin Research assistant for Zotero 10. Chat with your library, run federated scholarly search, RAG, OCR, systematic reviews, and manage cloud storage. Includes standalone MCP, Agentic capabilities, and skills library.
MIT ★ 97 -
MCP server for Mathpix OCR. Convert images, PDFs, and DOCX/PPTX to LaTeX and Markdown from any LLM/agent, over stdio or Streamable HTTP.
MIT ★ 2 -
Scanned-document workbench that AI agents can operate: clean, OCR, annotate PDF and DjVu, with a local MCP server. macOS, Windows, Linux, Web.
MIT ★ 5 -
MCP server for Lupa — let your AI agent search your Mac's documents by content (PDF, DOCX, HWP, OCR'd scans). Korean-aware, ~30ms, fully local.
MIT -
Privacy-first, self-hosted PDF toolkit with a built-in MCP server. Merge, compress, and OCR documents (including Russian) through a web UI or AI agents. One Docker container.
MIT -
MCP server that turns any video — YouTube, Instagram, TikTok, Loom, X, Vimeo, direct URLs, local files — into transcripts, key frames, OCR text, and metadata for AI agents.
npm MIT ★ 124 -
An MCP server that lets AI agents query your document archive instead of uploading it — with OCR, hybrid search, and cited local-AI answers in a private, Mac-native app.
MIT ★ 3 -
Windows desktop UI context picker for AI agents: select windows, UI elements, or screen regions with hover highlighting, UI Automation, screenshots, local OCR, and user-approved MCP output. Works with OpenAI Codex, DeepSeek Harness, and other MCP-compatible clients.
MIT ★ 2 -
MCP server that converts documents to Markdown for AI agents: PDF with layout, tables and OCR, DOCX, PPTX, XLSX, HTML, scans. docling in Rust, MarkItDown's convert_to_markdown tool.
pypi MIT -
KYC MCP server, eKYC MCP server, KYC onboarding MCP, remote KYC onboarding MCP, and identity verification MCP for AI agents: face comparison, selfie verification, face liveness, deepfake screening, document OCR, ID card OCR, Chinese ID card OCR, bank card
npm MIT -
Recordings in. Agent-ready evidence out. Local-first MCP for transcripts, keyframes, OCR, speakers, and wall-clock evidence.
MIT ★ 34 -
Agent-skillsgithub.com/Captcha-Sonic/Agent-skillsupdated this year
Solve any CAPTCHA with AI — reCAPTCHA, Cloudflare Turnstile, popular captcha, Geetest, AWS WAF, TikTok, OCR. Claude Code plugin + MCP server for AI agents.
MIT ★ 3 -
droidpilotgithub.com/youichi-uda/droidpilotupdated this year
Stable Android device automation for AI agents. Uses Accessibility Service + MCP protocol — no ADB, no OCR, just native OS APIs over WiFi.
MIT ★ 10 -
Open-source retrieval for agentic RAG: document search with exact citations on your laptop or servers, with OCR, CLI, MCP, and TypeScript.
npm AGPL-3.0-only -
computer use for ai agents on omarchy and hyprland: screenshots, mouse, keyboard, ocr, accessibility tree and window control. a claude code plugin and mcp server.
MIT -
Pay-per-call APIs for AI agents over x402: search, page reading, cited answers, PDFs, OCR, sanctions screening, company records, domain, email and SEO checks. USDC on Base, from US$0.01 a call.
MIT -
Agent Library personal knowledge base with a search and retrieval system MCP Server and CLI. Preconfigured local models and support for Text, images, OCR/vision, and code embeddings. Use over MCP with (Claude, Cursor, Codex, ChatGPT, etc)
Apache-2.0 ★ 34 -
Zeptapno update data
Operate a real iPhone from AI agents: screenshots, OCR, taps, swipes, typing and opening apps. Requires the Zeptap Mac app running with an iPhone connected over USB and paired over Bluetooth. Local stdio only, macOS 14+. Paid Mac app, first hour free. https://zeptap.com
smithery -
video-kyc-hackathongithub.com/himanisharrma/video-kyc-hackathonupdated this year
Real-time Video KYC (VCIP) — AWS Rekognition liveness, Textract OCR, Comprehend NLP, WORM audit trail + a 6-tool MCP server ('KYC Orchestrator') with PII masking and human-gated verdicts. Full Claude Code + Codex agent harness included.
-
document-blur-detector-ClarityGategithub.com/Manula-Fernando/document-blur-detector-ClarityGateactive
Ultra-lightweight (<60MB, Zero-GPU) Document Quality & Blur Shield Agent for KYC, NICs, Passports & Cheques before OCR (iDAX). With Multilingual Guidance & MCP Server.
Apache-2.0 -
Self-hosted web scraper, crawler and document parser for AI agents: URLs, PDFs with OCR and Office files to clean Markdown/JSON. Firecrawl- and Jina-compatible API, MCP server, agent endpoint, Python/Node SDKs. Single Rust binary.
npm Apache-2.0 ★ 274 -
Lets AI agents see and control a desktop GUI: screenshots, OCR, mouse, keyboard, windows and accessibility-tree UI elements. One-shot JSON CLI or MCP server. Windows 10/11, and Linux on KDE Plasma 6 or GNOME 45+ (Wayland).
MIT ★ 1 -
Universal Native Model Context Protocol (MCP) Server Hub bundling 19 production-grade AI agent tools: Web-to-Markdown, B2B leads, company deep-enrichment, real-time web search grounding, semantic cache, invoice OCR, headless screenshot, prompt shield fire
npm MIT -
MCP server that reads documents for AI agents: OCR of PDFs, scans, photos and Office files at two quality levels (Document Intelligence, or reconciled per page by a vision model). Signed-URL uploads, batch jobs, Entra auth, one Docker image.
MIT -
Local MCP server that gives any AI agent safe desktop control on Windows, macOS and Linux. Reads the screen as text and structure (accessibility + OCR) and acts by element name, with screenshots as the last resort. Multi-monitor aware, verifies every action, one safety gate.
npm MIT ★ 407 -
MCP server for VRoid Studio. Lets AI agents create and edit characters in the editor — choose presets, tune sliders and colors, frame the camera, and save. Drives the editor's own API through a BepInEx/IL2CPP plugin: no mouse or OCR automation; works with the window minimized
MIT ★ 13 -
Windows Computer Use for AI Agents. Both a tool (22 MCP tools for click, type, screenshot, OCR, UI inspection) and an agent (autonomous mission engine, macro recorder, intent-based discovery, event watchers). Built with opencode (DeepSeek V4). Ships as MCP server, web UI, and Tauri desktop app.
MIT ★ 53 -
Open-source test automation for Android, iOS, web, smart TV (Android TV, Tizen, webOS) and Bluetooth (BLE) devices. Write tests as CSV/YAML, record them live, or let an AI agent build them over MCP. Self-healing locators: XPath, text, OCR, image.
pypi Apache-2.0 ★ 99 -
Repairs Japanese dates broken by OCR (今和→令和, 曰→日, O→0) only when the calendar confirms the result. Stdlib Python, MCP server, MIT.
MIT -
FastMCP server providing advanced OCR capabilities with current state-of-the-art models (DeepSeek-OCR, Florence-2, DOTS.OCR, PP-OCRv5, Qwen-Image-Layered decomposition), WIA scanner control, and multi-format document processing for PDFs, CBZ comics, and images.
MIT ★ 25 -
Repair Japanese dates broken by OCR (今和→令和) and keep only fixes valid on the calendar.
pypi MIT -
本地优先的多模态视觉 MCP —— 为无视觉 LLM（DeepSeek、GLM 等）提供可自定义端点的看图 / OCR / 文档解析能力。
MIT -
MCP server for the Papyrio PDF API: compress, convert, merge, split and OCR PDFs from an assistant.
npm MIT -
Local macOS MCP server for screenshots, OCR, scrolling capture, annotations and screen recording, part of the Sightline app
-
Resilient streaming PDF table extraction, high-res page rasterization, AcroForm flattening, and OCR bbox dewarp repair.
MIT -
Novagithub.com/bigduu/Novaactive
Cross-platform Rust computer-use MCP server for macOS and Windows: AX/UIA, screenshots, OCR, pointer and keyboard input, clipboard, stdio, and HTTP.
MIT ★ 23 -
Local-first MCP server that turns video into queryable context with transcripts, keyframes, OCR, and semantic search.
MIT ★ 3 -
A MCP bridge that gives AI assistants real Windows GUI control: UIA/OCR/CDP element snapshots, click/input/verify loop, occlusion-safe window capture.
★ 2 -
Persistent JavaScript sessions for macOS computer use. Two MCP tools, native Accessibility, screenshots, OCR, and background-first interaction.
MIT ★ 1 -
Let Claude watch videos: timestamped transcript, scene-selected keyframes, exact frames and OCR. Local, no API keys. MCP server + skill + Claude Code plugin.
MIT ★ 3