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.
-
vision-mcpgithub.com/okysu/vision-mcpupdated this year
MCP server providing multimodal vision capabilities via OpenAI-compatible APIs
npm MIT -
kimi-vision-mcpgithub.com/ShaLuuFPS/kimi-vision-mcpupdated this year
kimi-vision-mcp — 让 Claude Code 拥有视觉能力,通过 Kimi Vision API 理解图片内容(截图/图表/照片/UI)
MIT ★ 1 -
vision-relay-mcpgithub.com/zhoucoolboy/vision-relay-mcpupdated this year
🖼️ Single-entry MCP server that gives Claude Code vision — relays local images to Anthropic/OpenAI-compatible vision APIs.
MIT ★ 1 -
reasonix-vision-mcpgithub.com/lxymp/vision-for-reasonixupdated this year
Vision Bridge MCP Server — 让 DeepSeek(纯文本模型)通过 MCP 调用视觉模型来「看到」图片。支持 SiliconFlow / OpenAI 兼容 API,含图片压缩功能。
npm MIT -
MCP server providing OCR capabilities to LLMs via Google Cloud Vision API - Read scanned PDFs, handwritten text, and images with any orientation
pypi MIT ★ 1 -
vision-mcpgithub.com/weekitmo/vision-mcpupdated this year
MCP server for image understanding through OpenAI-compatible vision APIs. To provide image recognition capabilities for those large models that do not support Multimodal.
MIT -
claude-vision-mcpgithub.com/EDMOK/claude-vision-mcpupdated this year
MCP server empowering text-only LLMs with image analysis via any OpenAI-compatible vision API
MIT ★ 5 -
mcp-local-visiongithub.com/jshsakura/mcp-local-visionupdated this year
Give text-only LLMs (GLM, DeepSeek, ...) eyes. MCP server that reads images and PDFs with a vision model on your own GPU — no API keys, nothing leaves the machine.
-
Local OCR and on-device screen reading for MCP clients (Claude, Cursor): read any window, locate elements to click, assert what's on screen. Apple Vision, no API keys, nothing uploaded.
npm MIT -
moondream-api-mcpgithub.com/hmdlohar/moondream-api-mcpupdated this year
MCP server for Moondream AI vision API with caption, query, detect, point, and segment capabilities
npm MIT -
@iflow-mcp/mcp-screenshot-website-fastgithub.com/just-every/mcp-screenshot-website-fastupdated this year
Fast screenshot capture tool for web pages - optimized for Claude Vision API
npm MIT -
MCP server that lets AI agents analyze, describe and read text in images through the free-first three.ws vision API.
NOASSERTION ★ 10 -
mcp-ocr-glmgithub.com/iwanro/mcp-ocr-glmupdated this year
Free OCR and vision for any MCP-capable AI coding client, powered by GLM-4.6V-Flash. Drop-in MCP server, no API key required.
MIT -
sight-mcpstale
MCP server for AI-powered image and video analysis - supports OpenAI, Claude, and multimodal vision APIs with local file and URL processing
npm MIT -
AdaTile-MCP: high-resolution image adaptive tiling MCP server for DeepSeek vision model (deepseek-v4-flash-vision-exp). L1-L6 pipeline (fastpath, saliency, adaptive tiling, Files API assembly, streaming VLM, rule-based merge) + eval harness. Setup: clone, run setup.bat, add your agent MCP config -> see README.
MIT -
Video frame extraction & composition service for AI vision pipelines, extracts frames via FFmpeg and composes them into 1×1/2×2/4×4 grid sheets for multimodal LLMs. REST API, Web UI, and MCP server over one core engine.
Apache-2.0 ★ 2 -
Local-first vision-language pipeline inspired by VL-JEPA. Compress images, text, conversations, and RAG documents locally via Ollama before sending to any LLM API. Deterministic, fact-preserving token optimizer plus MCP server, FastAPI server and video processing.
pypi Proprietary -
MCP server for image recognition using vision APIs (Anthropic, OpenAI, Cloudflare Workers AI)
pypi MIT -
MCP server for xAI’s Grok API with Web/X search, vision, image/video generation and file support
MIT ★ 56 -
OpenAI-compatible API over your signed-in ChatGPT and Gemini web sessions: chat, vision, image generation, and design tools that make website sections, app screens and brand images. MCP server, dashboard, embeddable widget.
MIT ★ 6 -
MCP server for the xAI Grok API — chat and agent loop, vision, image generation, text-to-speech, transcription
MIT ★ 3 -
Specialized MiniMax Model Context Protocol (MCP) server designed for coding-plan users, featuring AI-powered search and vision analysis APIs optimized for code development workflows
MIT ★ 101 -
chromium-screenshotsgithub.com/samestrin/chromium-screenshotsupdated this year
Vision AI "Cortex" for Agents. A Playwright-based MCP Server & API that captures screenshots with ground-truth DOM extraction and full auth state injection. Containerized.
MIT ★ 2 -
I am not a robot — a captcha solver for macOS: local model, Gemini vision, a stealth browser for checkbox captchas, RuCaptcha as the last resort. 2Captcha-compatible API and an MCP server.
MIT ★ 49 -
sandbox-agentgithub.com/eduresser/sandbox-agentupdated this year
LangGraph agent with Docker-based sandboxed code execution (Python, Node.js, R). Hardened containers with persistent kernels, vision support, provider-agnostic LLMs, cross-session file transfer, and session GC. Available as CLI, MCP server, REST API (Aegra), and Streamlit UI.
MIT -
roasr-mcpgithub.com/RoASr-com/roasr-mcpupdated this year
MCP server for RoASr — connect Claude, ChatGPT, Codex or Cursor to Meta / Google Ads / Yandex Direct ad-audit findings, KPIs, competitor spy and Vision creative analysis. Dependency-free client over the public /v1 API.
MIT -
MCP server for per-customer PDF document management and RAG. Ingests text, scanned and chart-heavy PDFs (PyMuPDF + Tesseract + vision model), isolates each customer in its own Qdrant collection, and returns cited answers with deep links to the original PDF page. Local-first with Ollama, or any OpenAI-compatible API.
MIT -
Harness de operaciones de ciberseguridad sobre Trend Micro Vision One: agentes de IA que hacen triage, investigan IOCs, ejecutan la contención y documentan la alerta. Incluye el servidor MCP (FastMCP) que expone su API, VirusTotal y Slack como herramientas. Ninguna acción de respuesta corre sin autorización humana, garantizada en código.
Apache-2.0 ★ 1 -
vision-mcpgithub.com/visianlee/vision-mcpupdated this year
解决 Claude Code + DeepSeek V4 不能看图 - 6免费Provider多模态视觉桥接 MCP Server
MIT ★ 4 -
mcp-vision-servergithub.com/joshsssn/mcp-vision-serverupdated this year
MCP server that exposes image analysis tools to LLMs via the Model Context Protocol.
MIT ★ 1 -
chart-vision-mcpgithub.com/nessos666/chart-vision-mcpupdated this year
Local TradingView chart analysis via OpenCV + Tesseract OCR. No API keys, no GPU, runs fully offline — MCP server with color/level/zone detection.
MIT ★ 3 -
An MCP-compatible weather server that provides `get-alerts` and `get-forecast` tools using the National Weather Service API, with optional email notifications.
npm MIT -
ForgeSyte is a modular AI‑vision MCP server
★ 1 -
A multi-interface (REST and MCP) server for automatic license plate recognition
pypi MIT -
Card recognition API for Pokémon, One Piece, Lorcana, Yu-Gi-Oh! and Magic: photo in, JSON with set, number, Cardmarket ID and prices out. Examples, MCP server, OpenAPI.
npm NOASSERTION -
A multi-interface (REST and MCP) server for automatic license plate recognition 🚗
pypi MIT ★ 27 -
nutrients-mcpgithub.com/ericblair1903/nutrients-mcpupdated this year
MCP server giving AI assistants food image & text nutrition analysis — calories, macros, vitamins, minerals, allergens. Powered by TastyAPI.
MIT -
AI-powered bug detection and auto-fix pipeline. A user reports a bug through a Chrome extension capturing a screenshot, DOM snapshot, and voice description and an autonomous agent diagnoses the issue, fixes the code in a sandboxed environment, commits the fix, deploys it, and closes the GitHub issue.
-
nvidia-mcpgithub.com/holm-digital-io/nvidia-mcpupdated this year
MCP server that gives Claude, Cursor and any AI agent free access to 100+ NVIDIA-hosted models (Nemotron, Llama, GPT-OSS, DeepSeek, vision, embeddings) with automatic task routing and parallel multi-model fan-out.
MIT ★ 2 -
Rewind your desktop. Ask what you were doing, circle what confuses you, approve what Enheduanna does. On-device screen memory, circle-to-ask vision, and governed actions, with no context-feeding by you. Private by design.
MIT ★ 7