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.
-
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 -
A PaddleOCR-powered MCP server: text, table, handwriting and formula recognition with LaTeX output — pip install ocr-mcp and plug it into Claude Desktop.
MIT ★ 1 -
Vision pipelines on a Hailo-8L M.2 NPU: MCP server (SCrFD face detect ~311 FPS, PaddleOCR v5, TinyCLIP embeddings), Spanish-aware OCR correction, measured multi-phase OCR quality program, and the Linux kernel VDMA stability patch. 100% on-device inference.
MIT -
hybrid-visiongithub.com/YashasviAsthana/hybrid-visionupdated this year
Hybrid Vision is a local "Visual Memory" server for AI Agents (Claude, Cursor, etc.). It acts as a Semantic Recorder that efficiently captures, indexes, and lets agents query your screen history without sending pixels to the cloud. Think of it as "Open Recall" for your AI assistant - running entirely on your local hardware.
MIT ★ 1