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.
-
RCLL — self-hosted shared memory for a team of AI agents. Rooms, hierarchical L0-L3 depth, MCP. The read path never invokes a language model. We publish the measured cost of one shared store versus ten private ones. Fork of vectorize-io/hindsight (MIT).
npm MIT -
Open-source, self-hosted AI platform for teams. Run private AI chat, manage Ray/vLLM GPU clusters, and connect apps and agents through an OpenAI-compatible LLM gateway and MCP.
Apache-2.0 ★ 2 -
Private, offline chat client for uncensored local LLMs. Zero dependencies, 228 tests, MCP server. Node + Ollama.
MIT -
MCP server that hands private-data or large-output work to a local LLM: the cloud model gets sanitized digests and stable placeholders, the material never leaves your network
MIT ★ 1 -
@siriusprotocol/mcpgithub.com/bdnwog/sirius-mainupdated this year
MCP server that gives an LLM agent a funded, spending-limited, private Sirius account — balance, private transfers, x402 paid fetch, quotes and swaps, over stdio
npm MIT -
A private, local memory layer MCP server for your LLMs
MIT ★ 6 -
Private, encrypted on-device memory for macOS: captures what you read, serves it to any LLM over MCP, and finds the loops you left open.
AGPL-3.0 -
sigilgithub.com/cdrn/sigilactive
Claude can sign, but never see. Local signing daemon + Claude Code integration that keeps private keys out of LLM context.
npm Apache-2.0 ★ 3 -
Fully local AI stack in Docker Compose — Ollama LLM inference, SearXNG private metasearch, Firecrawl scraping + real browser automation, and Langfuse observability behind one authenticated gateway with an MCP tool interface. No cloud calls, your data stays on your hardware.
MIT -
cdvgithub.com/azank1/cdvupdated this year
Conservative Dual-Verify — judging AI agent work with a deterministic floor + LLM critic (stricter score wins) and Bayesian stopping. Research preview, frozen at v1.0.1; active development continues privately. MIT.
pypi MIT ★ 4 -
Persistent AI memory with crossover — your AI at home shares the same memory as your AI at work. What one LLM learns, all your AIs know. Memory + PKM for Claude, GPT, Gemini & local models. MCP-native. (Engine stays private.)
npm MIT -
Local-first memory engine for AI-agent teams — private/team/project ACL, associative recall, federated sync, and a built-in MCP server. One SQLite file, no LLM required. Apache-2.0.
Apache-2.0 ★ 9 -
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.
-
Self-hosted multi-agent AI infrastructure: an MCP server exposing 48 tools for agent-driven infrastructure management, a multi-provider LLM gateway over commercial and self-hosted models, and a PostgreSQL knowledge graph with vector search. Architecture notes; source is private.
-
style-memory-mcpgithub.com/hexingyuofficial/style-memory-mcpupdated this year
A tiny local MCP server that learns how you talk — catchphrases, dialect, tone — without storing private memories. Local-first, no LLM calls.
MIT ★ 2 -
Connecting public and private LLMs to real-world humanoid robots via an MCP Server. Starting small, with LEGO Mindstorms and Hiwonder AiNex kits, and evolving toward fully autonomous, household-capable humanoids like Noetix Bumi.
-
A private AI finance copilot for investors: ask about your portfolio in plain language, get figures worked out in code with a trace of every answer, and keep your history locked with your own PIN. FastAPI, LangGraph and a local LLM via Ollama.
-
capix-mcpgithub.com/capix-protocol/capix-mcpupdated this year
Capix MCP Server — deploy and manage private LLM instances, compute, websites, and verified workloads on the Capix network from any AI coding agent that supports MCP.
npm Apache-2.0 -
AIDEFEND MCP is a local-first AI Security Defensive Assistant that brings the full AIDEFEND countermeasure library into your environment and turns static knowledge into actionable protection for LLMs and agentic AI systems — privately, securely, and on-device.
MIT ★ 18 -
Give any LLM a private, self-hosted brain on your NAS: categorized memory & skills, your tools, devices, email, calendar & webhooks, scheduled jobs that wake your Claude or run on ChatGPT, an API key or a local model, an encrypted vault, roles & per-user areas — single-user or multi-user, all through one MCP connector.
oci Apache-2.0 ★ 11 -
a local, private life dashboard — calendar, appointments, achievements, todos — that an llm can read and write via mcp. vanilla, stdlib-only, no cloud.
MIT -
An AI-powered employee assistant integrating Groq LLM and RAG via MCP for Claude Desktop. Features include secure employee onboarding, admin IAM for authentication, PDF-based company policy management, and seamless internal query access for employees through a private AI interface.
MIT ★ 2 -
A lightweight, robust Model Context Protocol (MCP) server built with Node.js and TypeScript to seamlessly connect LLMs and AI agents (Claude Desktop, Cursor, Antigravity, Cline) directly to GitHub repositories (public and private).
MIT -
Open-source, local-first long-term memory for AI coding agents: one MCP tool (recall) over a private local SQLite vault — zero cloud, zero vectors, zero background LLM calls, ~30 ms lookups. Apache-2.0.
NOASSERTION ★ 1 -
Fast, private, on-device dictation for Windows, Linux, and macOS. Hotkey to speak, text appears in any app. Whisper/Parakeet STT, meeting transcription with speaker labels, local LLM rewrites, MCP server for Claude/Cursor. Rust + Tauri, fully offline.
MIT ★ 1 -
devonbike.overfit.mcpgithub.com/devonbike/overfitupdated this year
A dependency-free MCP (Model Context Protocol) stdio server for the in-process Overfit LLM runtime — typed JSON-RPC 2.0 contracts with source-generated System.Text.Json serialization, no SDK dependency, no reflection, Native-AOT verified. Exposes local, zero-egress tools to MCP hosts (Claude Code, Claude Desktop, IDEs): ask a local GGUF model, query private documents with RAG citations, transcribe audio with Whisper — all in pure .NET on the CPU. Dual-licensed (AGPL-3.0-or-later / commercial).
nuget other -
Private, offline speech-to-text for Linux: push-to-talk dictation into any app, meeting transcription with speaker diarization, and searchable history. Whisper, Parakeet and a local LLM on your GPU (Vulkan) or CPU. Native Omarchy and Hyprland integration, Wayland, MCP for AI agents.
GPL-3.0 ★ 2 -
@thryx/mcp-servergithub.com/lordbasilaiassistant-sudo/thryx-mcp-serverupdated this year
ThryxProtocol v3.1.1 MCP Server — gasless AI agent launchpad + trading platform on Base. 21 tools wrapping the thryx.fun HTTP API. Server-managed wallets, paymaster gas, free LLM-driven autotrader. No private keys, no signing, no API bills. Auto-registers
npm MIT -
MTA_Plugingithub.com/GRU953/MTA_Pluginupdated this year
Give any AI assistant a private, token-free memory of your files — Claude, Gemini, Cursor, VS Code, Windsurf, Codex & Grok. Converts attachments (PDF/Office/EPUB/archives) to Markdown and builds a searchable knowledge-graph memory to recall from. 100% local, deterministic, model-free (no LLM/GPU). Incremental & portable; English + Bengali. MIT.
MIT ★ 1 -
Cardano MCPno update data
<p align="center"> <img src="https://cdn.simpleicons.org/cardano/0033AD" width="80" /> </p> <h1 align="center">Cardano MCP Server</h1> A **Model Context Protocol (MCP) server** for interacting with the Cardano blockchain from AI agents and automation systems. This server exposes wallet-aware tools for: - Submitting transactions - Fetching addresses - Reading UTxOs - Fetching balances - Resolving ADAHandles - Checking stake delegation ## Secrets 1. Copy the `.env.example` file to a new file `.env` 2. Specify your seed-phrase (`SEED_PHRASE`). Your seed-phrase is NEVER exposed to LLMs, and is only used for local wallet connection. 3. Configure the wallet connection details. Specify a Blockfrost Project ID (https://blockfrost.io/), or a Kupo & Ogmios URL. ## Docker Installation Build the container ```bash docker build -t cardano-mcp . ``` Run with your secrets ```bash docker run -p 8000:8000 \ -e PORT=8000 \ -e SEED_PHRASE="lizard,llama,frog..." \ -e BLOCKFROST_PROJECT_ID="your_blockfrost_key" \ cardano-mcp ``` ## Manual Installation Clone this repository ```bash git clone https://github.com/IndigoProtocol/cardano-mcp.git ``` Install dependencies ```bash cd cardano-mcp npm install ``` Build & Run ```bash npm run dev - or - npm run build npm run start ``` ## Connecting The URL for the Cardano MCP server will be hosted at http://localhost:8000/mcp ### Claude Desktop Configuration Add to `~/Library/Application Support/Claude/claude_desktop_config.json`: ```json { "mcpServers": { "cardano": { "command": "node", "args": ["/path/to/cardano-mcp/dist/index.js"], "env": { "SEED_PHRASE": "lizard,llama,frog..." "BLOCKFROST_PROJECT_ID": "mainnet123..." } } } } ``` ## Disclaimer By using this Cardano MCP Server and all related tools and technology (“MPC”), you acknowledge and agree that (i) your use of decentralized finance, including MPC and/or AI agents that you empower to manage your digital assets, involves various significant financial risks, including but not limited to, the risk of financial loss caused by MPC design or instructions, impermanent loss, and changes in digital asset prices; and (ii) that you are solely responsible for all MPC actions and transactions, and for securing your seed phrase, private keys, and environment configuration.
smithery -
Query Wikipedia ZIM archives offline with a hybrid RAG system using Llama.cpp. Build local indexes for private, internet-free knowledge retrieval.
-
PrivateGPT with MCP Server integration - AI-powered document QA with AI agent compatibility
Apache-2.0 -
Offline RAG app that explains Indian tax, bank and government notices with a local LLM. Code verifies every deadline, amount and quote against the source text. Includes an evaluation, audit log and read-only MCP server.
-
Policy-gated local SQLite MCP server that lets AI agents query sanitized database snapshots without exposing protected data.
MIT -
notion-private-api-mcpgithub.com/kirvigen/notion-private-api-mcpupdated this year
Unofficial Notion MCP server via the private API (token_v2) — full read/write, no integration token.
MIT ★ 4 -
MCP server for Netskope's Zero Trust Network Access Platform management through LLM's
npm ISC -
fundos-mcpgithub.com/8vdx1/fundos-mcpupdated this year
MCP server for FundOS — connect Claude Code, Cursor, and any AI agent to your fund operations
MIT ★ 1 -
The Privacy Firewall for LLMs
MIT ★ 80 -
MCP server and private mode for coding agents on OffRouter private inference. Every model runs in a GPU TEE, and your prompts are encrypted to the enclave before they leave your machine.
npm MIT -
Sapo-MCPgithub.com/nguyennguyenit/Sapo-MCPupdated this year
Model Context Protocol server for Sapo.vn POS & e-commerce — 105 tools across 4 modes, stdio + Streamable HTTP, single-tenant Private App.
npm MIT ★ 13 -
finprodata-mcpupdated this year
MCP server for finprodata — verified PE/VC, fund managers, and finance professional data for AI agents
npm MIT -
MCP (Model Context Protocol) bundle server for Coherence Gateway - enables LLMs to interact with domain-specific OpenAPI tool bundles
npm UNLICENSED -
MCP server and agent tooling for AzerothCore WotLK: unified queries over DBC, live SQL, terrain and playerbots data
GPL-2.0 ★ 6 -
Stop AI agents leaking secrets: MCP server + library that finds crypto wallet keys (BIP39 seed phrases with checksum, EVM/Solana private keys, WIF, xprv), API keys (GitHub, AWS, Stripe, OpenAI, Anthropic, Slack, Telegram, npm, JWT, PEM) and PII (cards, IB
npm MIT -
MCP server for SEC 10-K/10-Q segment revenue, BDC non-accrual, and Form 4. Every number carries accession and an EDGAR URL. PyPI edgar-filings-mcp 1.1.3.
pypi MIT -
Private, self-hosted knowledge base for Windows with document and code indexing, cited hybrid search, persistent AI context, and MCP, REST and CLI integrations.
MIT -
finance-intelligence-mcpupdated this year
A production-ready Model Context Protocol (MCP) server for private personal finance management.
pypi MIT -
Durable, private, time-aware memory engine for long-running AI agents
npm MIT ★ 1 -
smriti-dbgithub.com/labskrishna/smriti-dbupdated this year
Durable, private, time-aware memory engine for long-running AI agents
npm MIT -
mcp-servergithub.com/auxen-ai/mcp-serverupdated this year
Auxen MCP server — provision and manage private AI model endpoints on dedicated GPUs via Claude, Cursor, or any MCP client
-
Clean up messy folders with a local AI. Private, undoable. CLI + MCP server.
MIT -
@hippo-core/coregithub.com/hippocore/hippocoreupdated this year
Private local memory shared by MCP-compatible AI agents
npm MIT -
clinical-raggithub.com/stel-nik/clinical-ragupdated this year
Private on-prem RAG system for clinical document Q&A. Chunks, embeds, and retrieves using local models only. No data leaves the network. Includes an MCP server for AI agent integration and a private Llama3.1 agent. Built for regulated environments: FastAPI, Qdrant, Ollama, Docker, Kubernetes.
MIT -
Run a private, offline AI browser agent in Chrome with 18 built-in tools and zero cloud dependencies.
-
Self-hosted shared task board for AI agents — claim work, submit results over MCP. Private by design.
pypi MIT -
One private, local memory your AI tools share. Cross-model, on-device, never lose a session.
npm MIT ★ 1 -
OpenMemorygithub.com/carlos-life/OpenMemoryupdated this year
Cross-application long-term memory for AI — local, private, open source. MCP Server that lets every AI app share your context.
MIT ★ 5 -
Validated, private, shareable knowledge-graph memory for AI — per-tenant, write-gated, PostgreSQL-authoritative, served over MCP.
AGPL-3.0 ★ 20 -
Private AI platform on k3s — Ollama/Qwen3 + Milvus RAG + MCP agentic tooling + Fleet multi-cluster management
★ 1 -
Local, private long-term memory for AI. Remembers corrections, explains why it believes things, works with any MCP app. No cloud, no API key.
pypi MIT ★ 1