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.
-
ArchiGen Connect - AI Agent Bridge for AEC | Connect AI to Rhino, Grasshopper and computational design workflows using MCP.
★ 1 -
One Design MCP: image, brand, privacy and PDF tools from one.design, run on your own computer.
npm -
The 1-to-1 Design DNA Extraction & Transmutation Engine — clone any website's computed design system into your stack. MCP server + CLI.
npm MIT -
Vision Studio - Touchless Computer Vision, Gesture Recognition, 3D Kinematics & MCP Server (design influenced by Material 3)
MIT -
shot-canvasactive
Shot Canvas on your computer: a design canvas you and your agent share, over MCP. No account - canvases are files here until you share them.
npm UNLICENSED -
Describe a circuit in plain words: a language model designs it and checks it with a real SPICE simulator (ngspice), on your computer.
AGPL-3.0 -
Textile Designer AIno update data
Run the Textile Designer AI design tools on image files on your computer: print-ready upscaling, seamless repeats (block, half brick, half drop, mirror, step), Super Scaler, background and watermark removal, Dress to Design, Colourways, Colour and Object Layering, Channeling screen separation, Vectorizer, 3D and Embroidery effects, Sketch to Design and more. Results are saved next to your input image. Needs a Textile Designer AI account with credits or a plan: https://www.textile-designer.ai. Sign in with the `login` tool (browser device code). Privacy: https://www.textile-designer.ai/privacy. Source and docs: https://github.com/ScientiaAI/textile-designer-ai-claude
smithery -
Open-source platform for autonomous AI agents conducting quantum computing & scientific discovery. Features epistemic experiment loops, circuit co-design, a multi-tool federated MCP layer, and an immutable provenance ledger. Discover — then know.
Apache-2.0 ★ 2 -
MCP server for the MakerPortal compute API — biquad filter design, room eigenmodes, LLM VRAM fit. A transport shim: every answer is computed by the same function that renders makerportal.ai.
npm MIT -
Advanced MCP server for intelligent computer use automation and AI agent operator control. Supports mouse and keyboard input, window-relative vision and coordinates, Retina display, and screenshot-based debugging. Designed for AI-driven workflows, app and
npm MIT -
Lets an AI assistant draw and edit 3D parts in Onshape, a computer-aided design program.
MIT -
@automyc/studioactive
Automyc studio: a local MCP server that renders, edits and reviews video and designs on your own computer for Claude Code, Codex and other AI apps.
npm UNLICENSED -
Standalone server designed to deliver market data insights to AI trading agents. It processes data from external brokers and computes structured information to help LLMs and other AI systems make more informed trading decisions using JSON RPC protocol
★ 4 -
A private voice studio for your computer: text to speech, voice design and voice cloning with open-source AI models (Qwen3-TTS). Built-in MCP server for AI agents.
Apache-2.0 -
tv-mcpgithub.com/smitkunpara/tv-mcpupdated this year
A high-performance FastMCP server for programmatically scraping TradingView market data. The server provides unified endpoints for retrieving historical price data, computing technical indicators, and extracting market-relevant news. It is designed for automated research workflows, trading systems, and data-driven analytics pipelines.
MIT ★ 5 -
OpenSCAD wrapped as an MCP server. Render .scad source with parameter overrides, run parametric sweeps, validate syntax, extract module signatures, and compute STL metrics (bounding box, volume, surface area). Designed so an agent can iterate on para
MIT ★ 3 -
KMJ CodeBridge connects ChatGPT to authorized computers and VPSs through one MCP plugin for secure coding, Git operations, and verified builds and tests. Designed for lightweight agents, resilient connections, and centralized licensing through KMJ Main Platform. Under active development.
Apache-2.0 ★ 2 -
hw-sw-bridgegithub.com/BearsSki/hw-sw-bridgeupdated this year
Peer-to-peer MCP server for two Claude Code agents to coordinate over an encrypted Tailscale tunnel. Built for Computer Architecture final project — HW agent (Windows/Vivado) and SW agent (macOS/Xcode) exchange messages to co-design an iOS-to-FPGA system.
★ 1 -
Free, offline ElevenLabs alternative. Clone and design voices, dub videos, transcribe, isolate vocals and make audiobooks, all on your own computer. Works with Claude Code, Codex and Cursor via CLI and MCP. MIT licensed.
MIT -
mcp-ui-analyzergithub.com/manhhaycode/mcp-ui-analyzerupdated this year
This is an MCP (Model Context Protocol) server for analyzing UI screenshots and design images, integrating with GitHub Copilot. The server uses local AI models (Gemma 3 12B via Ollama) combined with computer vision tools (YOLOv8, OpenCV) to detect UI bugs, identify components, extract color palettes, and generate implementation suggestions.
-
We > Ultrarich was designed from the outset for both human and agent reuse. It has a free, no-auth, computation-only API + MCP server that answers questions like “How much is a billion (or trillion) dollars, really?” with ready-to-quote figures, sentences, and them-vs-you ratios. It's built for public use and its results are licensed CC BY 4.0.
NOASSERTION -
zadeh.aigithub.com/deegitech/zadeh-netupdated this year
The AI companion to Zadeh.NET: turn LLM confidence into deterministic Execute/Confirm/Clarify/Reject decisions (ConfidenceGate), generate fuzzy rule sets from plain-language policy via any LLM (RuleSmith), compute adaptive similarity thresholds for RAG/semantic caching (AdaptiveThreshold), and expose fuzzy engines as MCP tools for AI agents. AI designs, fuzzy decides — zero hallucination at runtime. Uses only the .NET base library — no external dependencies. Licensed under AGPL-3.0 / Commercial dual license.
nuget AGPL-3.0-only -
Human factors research translated into design standards, code examples, and an MCP server for AI-assisted validation.
npm NOASSERTION ★ 5 -
An MCP server that scores how faithfully your UI clones a design — layout, color, content, typography, spacing & motion — and returns an actionable punch-list for the hill-climb-the-clone loop.
MIT ★ 2 -
AI-powered Rhino/Grasshopper MCP with ML auto-layout
MIT ★ 18 -
MCP server that builds Grasshopper definitions programmatically - declarative recipes to editable .gh files via component-GUID resolution.
MIT ★ 1 -
An MCP server for JupyterCAD that allows you to control it using LLMs/natural language.
go Apache-2.0 ★ 20 -
iflow-mcp-asmith26-jupytercad-mcpgithub.com/asmith26/iflow-mcp_asmith26_jupytercad-mcpupdated this year
An MCP server for JupyterCAD that allows you to control it using natural language.
pypi Apache-2.0 -
AI-powered biologics design campaign agent — multi-agent orchestration with BoltzGen, PXDesign, Protenix, and 200+ cloud tools. Antibodies, nanobodies, de novo binders, and beyond.
go MIT ★ 120 -
Objetos que conversan: Raspberry Pi, Vidiv + ElevenLabs, luces y MCP
MIT ★ 6 -
grasshopper-agent-bridgegithub.com/chmhechtl/grasshopper-agent-bridgeupdated this year
A closed act-measure-correct loop between an AI assistant and a live Grasshopper document: the assistant places components, then reads back the data, the wiring and the geometry to check its own work.
MIT ★ 1 -
modulorgithub.com/bcllcc/modulorupdated this year
Agent-native 2D drafting + 3D modeling kernel - the geometry layer for the agent era. JSON in, geometry out.
pypi MIT ★ 3 -
MCP server for Rhino 3D: 105 tools for AI-driven 3D modelling, analysis, diagnostics and Grasshopper integration. Use with Claude.
MIT ★ 3 -
Rhino MCP server: AI-assisted architectural modelling in Rhino 8 with Claude, ChatGPT, Codex, Gemini or local Ollama. 128 tools: massing, vaults, arches, drawings, materials, PDF tracing, plus intent validation that catches errors valid geometry hides. Protocol 5.1, no .NET SDK required.
MIT ★ 35 -
Automated quality analysis for Grasshopper definitions. Real-time lint checking and code quality scoring to improve AEC industry productivity. Includes MCP server for advanced analysis.
MIT ★ 2 -
A declarative DSL for parametric CAD and 3D modeling, with an MCP server, browser editor, and live WebGL previews. JavaScript monorepo published to npm.
npm MIT -
mcp-visiongithub.com/local-intelligence/mcp-visionupdated this year
Local-first MCP vision server with explicit arming, auditability, and cloud-optional design
MIT -
mosso-mcpactive
Make and edit animations from Claude Desktop, Claude Code and Cursor. Runs on your computer.
npm SEE LICENSE IN LICENSE -
gridproofgithub.com/gridproof/gridproofupdated this year
Spacing & grid QA for AI-generated UIs, in the agent loop. MCP server that audits computed geometry and returns fix hints.
MIT ★ 4 -
Free, open-source desktop attention prediction (Windows & macOS). Predict where people look in the first ~3s of free viewing. 100% offline, private ONNX inference (SUM, UNISAL, TranSalNet). Transparent research-backed metrics (AOI share, clarity). GUI, CLI, local HTTP API, and MCP agent server.
MIT -
A collection of Model Context Protocol (MCP) servers that give AI assistants, agents and IDEs access to the FlagQuantum SDK. Reference design: Qiskit/mcp-servers.
pypi Apache-2.0 ★ 1 -
Automate FreeCAD with AI: a C# agent tool for parametric CAD (PartDesign, sketches, pad/pocket/revolve, patterns, fillet, booleans) with STEP/STL/3MF export. MCP-style local bridge; works with ChatGPT, Claude, Gemini, Copilot or a local LLM through any OpenAI-compatible endpoint. Windows, Linux, macOS. One-click installer, no Python.
AGPL-3.0 ★ 1 -
🔍 Streamline your SEO research with free tools designed for AI-powered IDEs, enhancing analysis and compliance with third-party services.
NOASSERTION -
Untiring, token-efficient research collaborator: reads papers and patents, cites the paragraph, writes the paper, designs and simulates what it proposes.
pypi GPL-3.0 ★ 11 -
A declarative DSL and compiler for reproducible gene-editing design — with LLM agents and an MCP server so any AI agent can drive it safely.
pypi Apache-2.0 ★ 1 -
OpenAqua Harness Beta — an evidence-grounded research workbench for early-stage water-treatment train design with CLI, TUI, Web, Electron, MCP tools, durable jobs, and provenance.
MIT ★ 1 -
steal-this-stackactive
Reverse-engineer any live website's exact computed CSS variables, typography, animations, SVGs, and interactive UI components into a ready-to-code Tailwind starter or feed directly to AI coding agents via MCP.
npm MIT -
🕸️ 3- Distributed MCP-based platform for AI incident analysis, with structured communication and traceability to support AI governance, Responsible AI and ethics-by-design.
MIT ★ 1 -
nsaf-mcp-servergithub.com/ariunbolor/nsaf-mcp-serverupdated this year
The Neuro-Symbolic Autonomy Framework integrates neural, symbolic, and autonomous learning methods into a single, continuously evolving AI agent-building system. This prototype demonstrates the SCMA component, which enables AI agents to self-design new AI agents using Generative Architecture Models.
★ 3 -
@heinrichb/web-suitegithub.com/heinrichb/web-suiteupdated this year
Config-driven Playwright + visual-regression harness for any website. Route-catalog discovery, reusable check primitives (a11y, broken-links, perf-budget, computed-style, visual-baseline), flow runner, edge-case generators, CLI, MCP server. Target-agnosti
npm MIT -
Jean by Melad Labs: the bioinformatics & biotech suite as a remote MCP server. 279 tools to design, simulate, analyze and verify experiments from any AI assistant: cloning, primers, CRISPR, sequencing, qPCR, genomics, phylogenetics, proteomics, chemistry.
★ 1 -
An end-to-end AI-powered relevance and visual content-matching engine with MCP and RAG capabilities, designed to evaluate, rank, and pair imagery with textual context using multimodal embeddings and deep visual-semantic scoring
AGPL-3.0 ★ 1