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.
-
MemoGraphgithub.com/Indhar01/MemoGraphupdated this year
MemoGraph is an intelligent personal memory management system for Large Language Models (LLMs) that transforms markdown notes into a queryable, graph-based knowledge base. It enables LLMs to access contextually relevant information from your personal knowledge vault using hybrid retrieval techniques.
pypi MIT ★ 4 -
LLM-COST-TRACKER-MCP-Server-NPM-Package-github.com/dominiquekossi/LLM-COST-TRACKER-MCP-Server-NPM-Package-updated this year
LLM Cost Tracker is a complete, production-ready NPM package that provides accurate token-based cost tracking for Large Language Models (LLMs). It includes a built-in MCP Server, a flexible budget manager, multi-provider support, usage analytics, cost simulation, and an extensible architecture for storage and pricing.
MIT -
KnowledgeManager_MCPgithub.com/Big-ShahMir/KnowledgeManager_MCPupdated this year
A highly efficient local server that implements the Model Context Protocol (MCP). It acts as a bridge, providing Large Language Models (LLMs) with direct, tool-based access to a curated knowledge base built from local PDF documents.
-
mcp-excel-server is a project based on the Spring AI MCP framework. Its main function is to provide large language model (LLM) with the ability to read and write local Excel files with the help of MCP protocol.
★ 7 -
mcp-rag-server is a Model Context Protocol (MCP) server that enables Retrieval Augmented Generation (RAG) capabilities. It empowers Large Language Models (LLMs) to answer questions based on your document content by indexing and retrieving relevant information efficiently.
npm MIT ★ 36 -
nUR_MCP_SERVER is an intelligent industrial collaborative robot control middleware system built based on the MCP (Model Control Protocol) interface protocol. It realizes the natural language interactive control of industrial robots by integrating a large language model (LLM).
NOASSERTION ★ 8 -
FileAsist, a file organization agent running on small language model, on your local, reading you files, understanding them and organizing it based on your preferences.
★ 1 -
selenium-mcp-servergithub.com/mm541/selenium-mcp-serverupdated this year
It is a standalone Python application that exposes a comprehensive suite of Selenium browser automation commands as a FastMCP server. This allows a language model, AI agent, or any other client capable of calling tools to control a web browser, perform complex web scraping, and automate web-based tasks by issuing natural language commands.
-
Offline AI knowledge base: read, search and ask your ZIM library (Wikipedia and the Kiwix catalog) with a local or hosted model, and serve it to LLM apps over MCP.
pypi GPL-3.0-only -
This project automates email-based user registration for LLM apps using the Model Context Protocol (MCP). It acts as a simple backend that receives email addresses from users, registers them with an MCP-compatible LLM server, and returns the generated API key or access token.
Apache-2.0 -
medha-mcpgithub.com/tejzpr/medha-mcpupdated this year
Medha is a Model Context Protocol (MCP) server that provides persistent, git-versioned memory storage for LLM applications. It combines the power of Git's version control with graph-based memory associations, enabling LLMs to maintain long-term memory.
MPL-2.0 ★ 14 -
orchestrator-mcpgithub.com/cameronking4/orchestrator-mcpupdated this year
The Task Orchestrator MCP is a specialized state-management primitive designed to give AI agents "Executive Function." It allows an LLM to externalize its planning process, breaking complex goals into trackable, hierarchical tasks that persist across turn-based interactions.
npm -
This project implements a lightweight MCP (Model Context Protocol) server for ReportPortal using FastAPI. It pulls test metadata from ReportPortal and exposes it through a clean API for use in LLM-based tools, dashboards, or QA automation systems.
-
browser-mcpgithub.com/zzzhhhaaannnggg6/browser-mcpupdated this year
Vibe-coded browser-automation MCP server. Lets an LLM read, search, and traverse a private knowledge base via a persistent browser session — no vendor API key. Personal study project; the security questions surfaced while building it are turning into my undergraduate research interest.
-
Agent Trust Layer — rule-based discriminator engine for AI agent output verification. 47 dimensions, 9 check layers, 131 MCP tools. Zero LLM dependency. Checks what AI says before it reaches humans and says "no" when something is wrong.
npm MIT -
foragithub.com/net-forge/foraupdated this year
Fora is a CLI-first forum platform for autonomous AI agents. It provides a single Go HTTP server backed by SQLite, a CLI client for agents/admins, and an MCP bridge for tool-based LLM integration.
Apache-2.0 -
Airbnb Price MCP server: exposes a FastMCP tool that loads the trained scikit-learn pipeline to predict nightly listing prices, plus a helper prompt, shared SVG icon, and docs so LLM-based clients can call it reliably.
-
OpenAIModeration_MCP_APIgithub.com/turbotrail/OpenAIModeration_MCP_APIupdated this year
This project provides a comprehensive content moderation solution using OpenAI's Omni moderation model. It exposes a FastAPI service for HTTP-based interactions and a Model Context Protocol (MCP) server for LLM tool integration.
-
The MCP-RAG project is an AI agent framework based on the Multi-Context Protocol that connects various data sources (MySQL, documents, RAG search) with LLMs. It supports HTTP API, SSE chat, an MCP router, batch workers, and a React + Tailwind frontend.
Apache-2.0 ★ 10 -
obsidian-mcp-ultragithub.com/alch3mistdev/obsidian-mcp-ultraupdated this year
obsidian-mcp-ultra is a high-performance MCP (Model Context Protocol) bridge for Obsidian, exposing your vault as a structured, queryable knowledge graph for agentic workflows. It enables semantic retrieval, context injection, and bidirectional edits—turning your personal knowledge base into a live substrate for LLM-driven systems.
npm MIT ★ 1 -
WhatsApp Messaging with MCP & Ollama This project integrates PraisonAI Agents with Ollama (llama3.2) and a Go-based WhatsApp bridge. It allows you to send WhatsApp messages using natural language through a local REST API, powered by an LLM and custom MCP tools.
MIT -
An Agentic Retrieval-Augmented Generation (RAG) system leveraging LangGraph for workflow orchestration, MCP for tool integration, and PostgreSQL/pgvector for hybrid information retrieval. It dynamically routes queries across document management, database operations, local document search, and web search, with LLM-based relevance refinement to impro
Apache-2.0 -
An open-source MCP server that turns LLMs into condition monitoring assistants for reliability engineers. It gives your AI assistant evidence-based vibration diagnostics: machine fault detection, severity assessment and diagnostic reports, for faster predictive maintenance decisions. Also available as a Claude Code plugin.
pypi MIT ★ 114 -
Plugin-based macro stress testing over MCP. A person names an event; an LLM agent writes the scenario from a corpus of sources, and the person approves it. Models in levels, from the economy to industries and companies, fill in the paths and reprice positions; each model and asset type is a plugin. Output: loss quantiles, ES95, scenario price.
pypi MIT ★ 5 -
SEO Performanceno update data
Post-publish SEO performance MCP. Unifies **Google Search Console, Matomo, GA4, Microsoft Clarity, and AI-citation signals** per URL and emits a **verdict per post** (refresh / expand / merge / kill / double_down / hold) with deterministic reason codes and a 0-1 confidence score. Reporting only — never mutates posts. Works with any CMS via XML sitemap; optional Ghost Admin API integration for richer metadata. ## Tools - **posts.list** — discover posts via XML sitemap, JSON override, Ghost Admin API, or explicit URL list. - **posts.snapshot** — unified 30/60/90-day signal pull across every configured source for one URL. - **posts.decay_curve** — weekly GSC bucket + decay / plateau / growth classification. - **posts.verdict** — rule-based verdict engine: refresh, expand, merge, kill, double_down, hold. - **posts.refresh_brief** — markdown brief: verdict + reasons + numbers + suggested actions. - **posts.cite_loss** — list of LLMs that previously cited this URL but no longer do. - **cohort.report** — verdict engine across a cohort, ranked by verdict priority + confidence. - **gsc.quick_wins** — pages at positions 5-15 with low CTR, ranked by impressions. ## Prompts - **audit_cohort** — run cohort.report and generate refresh briefs for every actionable URL. - **find_quick_wins** — propose query-verbatim title and H1 rewrites for SERP underperformers. - **citation_loss_sweep** — find URLs that lost LLM citations and recover them. ## How it works Each source is best-effort: if its env vars are missing the field is omitted. Returns whatever is available. Optional DuckDB cache layer for historical comparisons. Platform-agnostic — works with WordPress, Hugo, Astro, Next.js, Webflow, Framer, Wix, Squarespace, Substack, Ghost, and any CMS that exposes a sitemap.
smithery -
Cross-vendor multi-model review committee that gates AI/agent-generated content before it ships — consensus-based hallucination defense. Zero dependencies, local-first CLI, MCP server, CI gate exit codes.
GPL-3.0 ★ 1 -
agent2kb turns everything your agents find into a living, cited Markdown wiki. An LLM compiles it in your own bucket; agents read it over MCP, people read it in Obsidian or any Markdown editor.
AGPL-3.0 -
A local knowledge base MCP server for **Odysseus**. It indexes folders of text documents into SQLite and lets Odie search them by keyword. It uses only the Python standard library, with no extra dependencies.
-
easymemgithub.com/zhangxiangliang/easymemupdated this year
An LLM wiki your coding agent writes — agent memory you can open and read. easymem stores it, searches it, and links it. Plain markdown on disk, no vector database. Runs with npx or as an MCP server.
npm NOASSERTION ★ 1 -
An open-source MCP server that turns any LLM into an Intelligent Tutoring System. 50 years of cognitive science, MIT licensed.
go MIT ★ 103 -
Run your own priced knowledge base for AI agents. One slot per question, the largest ALGO stake holds the answer, a bigger stake replaces it in the same transaction, and reads pay whoever held it. MIT, MCP tools, x402-shaped paid reads, Algorand TestNet.
MIT ★ 1 -
Open Source Implementation of Karpathy's LLM Wiki. Upload documents, connect your Claude account via MCP, and have it write your wiki !
Apache-2.0 ★ 1,771 -
The Open-Source Knowledge Base for AI Agents. Tell one agent once: Claude Code, Codex, Cursor, and Copilot all remember it, and Libro, an AI librarian, keeps it organized and current.
Apache-2.0 -
Vector search that ships as a file. Compile a corpus into one self-contained, hash-verified .pikelet — search it anywhere, or hand it to an LLM agent over MCP.
npm Apache-2.0 ★ 2 -
Verifiable memory for AI: knowledge only counts once a real test confirms it. Project memory across restarts, MCP server for Claude, local Rust + SQLite.
MIT -
Dexio for Cursor: a shared wiki your agents keep, over MCP, plus twelve skills for keeping it correct.
MIT -
alexandria-wikigithub.com/epappas/alexandriaupdated this year
Local-first single-user knowledge engine — accumulates gathered knowledge and exposes it via MCP to connected agents like Claude Code.
pypi MIT -
Shared, sourced memory for AI coding agents: ingest Claude Code/Codex sessions, keep confidential material out of git, serve it over MCP.
MIT -
A memory for your AI coding assistant. It keeps what you tell it and hands the right note back at the moment it matters. Runs entirely on your own machine.
oci GPL-3.0 ★ 5 -
No-code visual node editor for AI workflows and small local tools: describe each node in plain words, AI writes its code or prompt, wire nodes into a graph and run it offline with a local LLM. Local-first, no telemetry, no account. Source-available, free for noncommercial use.
NOASSERTION -
LLM-maintained, fully-cited personal wiki in Open Knowledge Format — with an MCP server so an AI can search and read it.
pypi MIT ★ 3 -
mcp-guardrailgithub.com/mini4ai4/mcp-guardrailupdated this year
An MCP server that gives an AI agent read access to a database without giving it the database — enforced by SQLite's authorizer, proven by an adversarial fuzz harness.
MIT -
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 -
Scores any public URL on whether AI agents can discover, understand, call, and trust it. Machine-payable via x402 — USDC on Base, no account.
MIT ★ 1 -
Dynamic routing for agent knowledge. Areas advertise when they're relevant; the agent routes before it reads. MCP server, git-backed, no database.
MIT ★ 4 -
Machine-consumable error knowledge base for open-source libraries — MCP server, REST API, and static dataset mapping every error message to the source line that raised it
MIT -
Memory for coding agents over MCP: stores every fix and finds it again when the problem returns. EU-hosted.
npm Apache-2.0 ★ 2 -
A block-graph store for LLM agents and the people working with them, with its daemon and MCP shell.
npm Apache-2.0 -
Team knowledge evaporates daily — pairing sessions, debugging context, architectural rationale lost to Slack. Distillery captures it at the point of creation, connects it into a living graph, and surfaces it conversationally. It monitors feeds, tracks what matters to your projects, and alerts you before you know to ask. A team brain that learns.
NOASSERTION ★ 25 -
An MCP server for markdown/Obsidian knowledge vaults that ships with its own eval suite: cited answers, honest abstention, index-level entitlements and a redaction guard, all CI-gated.
MIT -
Self-hosted alternative to Jira, Confluence and Redmine: issues, wiki and AI search that cites your own docs. Agents work in it safely via MCP.
AGPL-3.0 -
The open framework AI agents use to share knowledge with one another. Pack knowledge into a portable, verifiable Knowledge Image, pin it by digest, mount it locally over MCP, and prove what was used. Local-first, deterministic, no vector DB required.
npm Apache-2.0 ★ 108 -
Code-KBgithub.com/liuyi585851/Code-KBupdated this year
Retrieval & navigation service for a code monorepo + its docs: indexes source and docs into small, locatable "atoms" and answers with a strict cite-or-die policy. Agent-first, MCP-native + HTTP API.
MIT ★ 1 -
molinggithub.com/gojue/molingupdated this year
MoLing is a computer-use and browser-use based MCP server. It is a locally deployed, dependency-free office AI assistant.
Apache-2.0 ★ 340 -
lingtai-mcpgithub.com/edrc-shawn/lingtai-mcpupdated this year
Self-evolving knowledge management system: Obsidian vault + a zero-dependency MCP server (86 tools). Refines raw materials into structured knowledge, evolves its own rules, and runs 24/7 patrol automation
MIT ★ 1 -
A shared fix-pool for coding agents with a write-time poisoning shield — every post_solution is screened for prompt-injection and leaked secrets before it can reach a reading agent.
Apache-2.0 ★ 1 -
RuvNet Brain — a downloadable, source-grounded brain for Claude Code over Reuven Cohen's (rUv's) RuvNet stack: RuVector/RVF, Ruflo, AgentDB, RuLake, SPARC + 21 building blocks. Grounds Claude in real source via one MCP tool (search_ruvnet), so it builds with the stack instead of drifting off it.
npm NOASSERTION ★ 84 -
Resolver for AI agents. An agent works something out once: Genter remembers where the answer is written and reads it there again, so the answer is always current. Zero dependencies, no keys: CLI, MCP server and library.
npm Apache-2.0 -
docmind-mcpactive
Give your coding agent a document brain: point it at any folder, get instant search, budgeted LLM research, and persistent memory via MCP.
npm MIT -
deep-research-mcpgithub.com/pinkpixel-dev/deep-research-mcpupdated this year
A Model Context Protocol (MCP) compliant server designed for comprehensive web research. It uses Tavily's Search and Crawl APIs to gather detailed information on a given topic, then structures this data in a format perfect for LLMs to create high-quality markdown documents.
go Apache-2.0 ★ 27