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.
-
AI Learning Assistant MCP Server - Educational tools API powered by Cloudflare Workers. Includes Wikipedia search, translation, and more learning utilities.
MIT ★ 2 -
Minimal NestJS Streamable HTTP MCP server for learning resources, tools, deployment, and Codex integration
MIT -
kiro-learngithub.com/brendangeck/kiro-learnupdated this year
Continuous learning for Kiro agent sessions on AWS. Passive tool-use capture, LLM extraction, and synchronous context injection — inspired by claude-mem, aligned with Bedrock AgentCore Memory.
npm Apache-2.0 -
LinkedIn Learning API: Python + MCP quick-start on Apify. Call it from Python (uv) or as an MCP tool in Claude and Cursor. Returns structured JSON for course data, learner reviews, syllabus and instructors, no login needed.
-
ml-trainer-mcpgithub.com/abhijithb-r/ml-trainer-mcpupdated this year
An MCP server that trains and evaluates machine learning models through MCP tools
pypi MIT -
morphnetgithub.com/suswader/morphnetupdated this year
Transforms browser automation into reusable API tools. Uses computer use as discovery infrastructure to learn deterministic MCP tool calls from live site traffic.
pypi MIT -
DSH autonomous-learning plugin (v2): TypeScript engine, 11 native DSH tools, evidence-based verification, optional MCP adapter.
MIT -
Java employee-leave AI learning project using Spring AI, Gemini, RAG, tool calling, and an MCP server.
-
Learnlog MCP: ask your AI assistant what learning records actually show about help. Read-only tools for Learnlog records, QLM credential verification, and QLM's public misconception and learning-graph datasets.
npm Apache-2.0 -
A learning project for game-art workflows: .NET 8/WPF desktop tools, local image processing, durable generation tasks and Agent/MCP integration. 游戏美术工作台与工程学习参考。
MIT ★ 1 -
day10-Weather-mcpgithub.com/nyalamrithwik-oss/day10-Weather-mcpupdated this year
Weather Intelligence MCP Server - Production-ready weather tools for Claude Desktop. Part of 30-Day RAG & MCP Learning Journey
MIT -
quizz-mcpgithub.com/ThoBustos/quizz-mcpupdated this year
Learn while building. Useful tool to learn from your coding sessions/documents.
MIT ★ 3 -
Japanophile workstation: kanji/JLPT learning tools plus Japanese culture knowledge box. First -phile repo.
MIT -
McPortfolio: A Model Context Protocol server providing 9 specialized tools for LLM-driven portfolio optimization using natural language, covering mean-variance to machine learning approaches.
Apache-2.0 ★ 6 -
Persistent memory MCP server for AI agents — Rust, 19 tools, knowledge graph, Hebbian learning, episodic memory, contradiction detection, prospective triggers, Bayesian calibration, zero-config Docker setup.
npm Apache-2.0 ★ 30 -
Model Context Protocol server for Cognivia. Gives any AI agent learning diagnostics (error type, memory state, session report) as callable tools.
npm MIT -
Local-first learning tool: a concept-graph dashboard over plain files, with an MCP server so AI can weave lessons, tests and roadmaps.
npm MIT -
Governed, self-improving multi-agent swarm for 5G network operations on AWS: Bedrock agents, MCP tools behind AgentCore Gateway, Cedar policies, human approval, RL + federated learning on a digital twin. Finalist, AWS x FYUZ Breaking Barriers 2026 (team project).
-
Multi-agent pharmaceutical supply chain optimizer using OECD Pharma data (19.6M rows), World Bank financing, and reinforcement learning with MCP tool orchestration
-
@swarmrecall/mcpgithub.com/swarmclawai/swarmrecallupdated this year
Model Context Protocol server for SwarmRecall — exposes memory, knowledge graph, learnings, skills, pools, and dream tools to any MCP-compatible agent over stdio
npm MIT -
telegram-ai-agentgithub.com/abailey81/telegram-ai-agentupdated this year
Production-grade conversational AI platform — 13 intelligence engines, 6 deep learning models, Thompson sampling RL, 4-backend voice synthesis, and autonomous model improvement. 90+ MCP tools, 80+ REST endpoints.
Apache-2.0 ★ 3 -
workramp-mcpgithub.com/msdanyg/workramp-mcpupdated this year
MCP server for the WorkRamp LMS API — 30 tools that let Claude manage users, courses, learning paths, certifications & academies in plain English.
MIT ★ 1 -
Kubernetes, Helm, OpenShift, ROSA HCP and RHACS expertise as deterministic tools: audit, diagnose, cited search, learning. No AI inside: use it from Claude, OpenShift Lightspeed, any MCP client or the CLI.
pypi Apache-2.0 -
Permanent code intelligence layer. Learns what matters, expires what changed, proves what existed. Content-addressed graph with Merkle proofs, 25 extractors, 23 MCP tools. Gets smarter with use.
pypi Apache-2.0 ★ 24 -
MCP server for Cubite LMS, a hosted learning management system (SCORM, xAPI, certificates, cohorts). Run your LMS from Claude, ChatGPT, Gemini CLI or Cursor: 137 tools, OAuth sign-in.
MIT -
Robot learning stack for a real WidowX AI arm: demo capture, calibrated MuJoCo digital twin, Diffusion Policy with real-time chunking, π₀.₅/LeRobot in Docker, and LLM tool-calling control over MCP.
NOASSERTION -
ATLASgithub.com/web3guru888/ATLASupdated this year
Active-inference Training with Learned Adaptive Stigmergy — Pure Rust AGI framework, 21 crates, 565 tests, 28 MCP tools, BF16 GPU inference (15.4 tok/s OLMo-3-7B on A100), OpenAI-compatible API, ZK proofs. Zero external dependencies.
Apache-2.0 ★ 3 -
example-agent-tool-servergithub.com/mldotink/example-agent-tool-serverupdated this year
Learn how to deploy a custom tool server for AI agents with Ink MCP.
-
Ayakagithub.com/Kvyii/Ayakaupdated this year
MCP tools for Japanese learning
GPL-3.0 ★ 2 -
zsh-toolgithub.com/ArkTechNWA/zsh-toolupdated this year
Zsh execution MCP tool for Claude Code with yield-based oversight, NEVERHANG circuit breaker, and A.L.A.N. learning
pypi MIT ★ 5 -
Creativity and institutional-learning MCP for quant research orgs (28 tools, SQLite-backed)
-
first-mcpgithub.com/thegostisdead/first-mcpupdated this year
Built to learn how MCP tools work
-
foragegithub.com/isaac-levine/forageupdated this year
Self-improving tool discovery for AI agents. Agents find, install, and learn to use new MCP tools automatically.
npm MIT ★ 19 -
A server for managing contextual data in DICOM tools, supporting medical imaging and machine learning workflows.
LGPL-2.1 ★ 3 -
Behavioral memory layer for AI tools. Learns preferences and context across Claude, Cursor, Codex.
pypi -
engram-toolsactive
Memory for AI agent teams. Store, recall, and learn across tools, sessions, repositories, and teammates. mcprush.com-claim-engram-engram-mcp-03956b462406295db4640c05
npm MIT -
Veracity_MCPgithub.com/chadudell/Veracity_MCPupdated this year
MCP server for Veracity Learning Record Store (LRS) API — 20 tools for learners, courses, classes, launches, and analytics
★ 5 -
X-Automation-MCPgithub.com/nathanbenaiah/X-Automation-MCPupdated this year
A tool that learns your niche, writes posts, and publishes them on X through an MCP server.
MIT -
Can LLMs learn a programming language they've never seen? A from-scratch compiler for MiniLang, used as an uncontaminated benchmark where Claude, GPT and Qwen learn it from the spec and debug through a compiler-as-tool agent loop. MCP server included.
-
learnforge-mcpgithub.com/techshare101/learnforge-coreupdated this year
LearnForge MCP Server — AI-powered adaptive learning tools for Claude Desktop
npm MIT -
@piyushdua/engram-devgithub.com/pdgit12/engramupdated this year
Second brain for Claude Code — MCP server with 20 tools for code intelligence, learnings, and team orchestration
npm MIT -
orchestr8github.com/mrzadexinho/orchestr8updated this year
Agent coordination for Claude Code — shared memory, smart routing, and learning. 13 MCP tools, 90 tests, zero fluff.
npm MIT ★ 1 -
Claude Code plugin for Arma Reforger modding: MCP tools and skills to learn, build, and fix mods. Unofficial. Work in progress.
MIT -
Learns the tool-call workflows your AI agent repeats and replays them: 38% fewer tokens at 100% accuracy in benchmarks. Claude Code, Codex, MCP.
MIT -
openplan-mcpgithub.com/anomalyco/opencodeupdated this year
Waze for AI agents — plan, track, and learn from software projects. 3 tools, one job each.
pypi MIT -
A Model Context Protocol (MCP) server that enables Claude Desktop to interact with Google's Gemini AI - featuring 7 tools with Smart Tool Intelligence that learns from your usage patterns
MIT ★ 19 -
bodigi-mcp-servergithub.com/bobbiedigital2025/bodigi-mcp-serverupdated this year
bodigi-mcp-server is a shared MCP (Model Context Protocol) tool server that powers learning, automation, and adaptive intelligence across all BoDiGi applications.
MIT -
A dedicated MCP (Model Context Protocol) tool for github, The MCP (Model Context Protocol) Tool for GitHub is a Python-based solution designed to help manage and track machine learning model context directly within GitHub repositories.
★ 1 -
mcp-serversgithub.com/nishilfaldu/mcp-serversupdated this year
A small fleet of FastAPI tool servers (Canvas, Google Calendar, Outlook) that expose external APIs as LLM-callable tools — built while learning to wire MCP-style integrations into an agent
-
BitcoinCostBasisAigithub.com/aligneddev/BitcoinCostBasisAiupdated this year
Learning MCP and Microsoft Agent Framework and using F# to create a tool to help with Bitcoin Cost Basis for taxes with a LLM front-end
MIT ★ 1 -
Learning project: MCP server (tools, resources, prompts) and a Claude-powered CLI chatbot that uses it
-
A MCP server that will respond using tools and resources provided (locally). This is a learning purpose project, feel free to suggest changes and contribute.
Apache-2.0 -
A Model Context Protocol server with machine learning engineering tools: BPE token counting and cost estimates, BM25 search, paired bootstrap and McNemar evaluation comparison, JSON Schema validation, and prompt-injection and PII scanning.
MIT -
Local-first assistant that watches your Jira/Slack/GitHub and learns which tools actually solve which problems — an agentic loop graded mechanically, no LLM-as-judge.
NOASSERTION -
gentle-mcpupdated this year
GENTLE - Game Engine Navigation Tool for Learning & Exploration. MCP server for Unreal Engine and PyQt6 documentation.
npm MIT -
preflightgithub.com/preflight-dev/preflightupdated this year
✈️ 24-tool MCP server for Claude Code: preflight checks for your prompts, cross-service context, session history search with LanceDB vectors, correction pattern learning, cost estimation
MIT ★ 13 -
Seamark helps AI coding agents stop repeating mistakes by learning from Git history and code reviews. Repository lessons, MCP tools, and agent skills.
Apache-2.0 -
Vetono update data
93 agentic MCP tools + 49 specialist agents for every major AI CLI — self-learning, no API keys. Veto gives Claude Code, Codex CLI, Gemini CLI, Cursor, Windsurf, Zed, and JetBrains a council of specialist agents, persistent cross-platform memory, a self-learning router, CI/CD gates, and dependency-hallucination + decision-drift guards — all powered by your existing AI subscription. No separate API keys, no extra billing. **Highlights** - **Council debate** — multi-agent GREEN/YELLOW/RED verdicts on risky changes (LLM-backed via MCP Sampling, with deterministic fallback) - **Cross-platform memory & session handoff** — save on Claude, resume on Gemini or Codex; nothing needs re-explaining - **Self-learning router** — improves task routing from recorded outcomes - **Guards & gates** — CI gate, diff review, secrets scan, dependency-hallucination verification, decision-drift enforcement - **Compact mode** — advertise 9 tools instead of 93 and save ~16K context tokens; every tool stays callable **Links**: [GitHub](https://github.com/jigyasudham/veto) · [npm](https://www.npmjs.com/package/@jigyasudham/veto)
smithery -
better-git-standupgithub.com/bayquen/better-git-standupupdated this year
Dual-purpose MCP server for: 1) AI-enhanced git standup and, 2) learning how to implement custom MCP tools for agents
MIT ★ 1 -
alberta-tax-agentgithub.com/chiKeka/alberta-tax-agentupdated this year
Alberta business tax filing agent. Claude MCP plugin with self-learning capabilities. 47 tools across corporate tax, GST, payroll, and tax planning.