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. Each candidate is checked by a language model from its name, description and topics, and only actual MCP servers are listed (not clients, SDKs, lists or skills). Package descriptions, licences and links are read from each package's own registry (npm, PyPI, crates.io, NuGet, RubyGems; Go and Maven packages are only checked to exist), and packages the registry no longer has are removed. 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.
-
Ferramenta de consulta de bases de dados usando linguagem natural com agente inteligente
pypi MIT License -
Let an AI agent write to Postgres one statement at a time: dry-run first, current-value checks, row cap enforced in the transaction, masked snapshots. CLI + MCP server, no dependencies.
MIT -
HARNESS-DB: 1,256 LLM agent harnesses coded on 38 design dimensions with a verbatim quote behind every value; pre-registered PRISMA 2020 review; loader, explorer, MCP server
MIT ★ 3 -
npx launcher for db-conn-mcp: a dead-simple, self-hosted MCP server for safely querying your databases with AI agents. Runs the PyPI release through uv.
npm MIT -
DB_MCP_Servergithub.com/KISS-GG/DB_MCP_Serverupdated this year
DB_MCP_Server 是一个强大的数据库连接和操作服务,通过 MCP (Model Context Protocol) 协议为 AI 应用提供数据库访问能力。项目采用响应式编程模型(WebFlux),支持多种主流数据库,为 AI Agent 提供安全、高效的数据库操作接口。
MIT ★ 3 -
DB_mcpgithub.com/krishnarg04/DB_mcpupdated this year
DB_mcp provides AI Agent with access to relational databases via the Model Context Protocol, supporting operations like schema exploration, SQL execution, and data visualization. Built for integration with tools like IDEs or CLI Desktop
-
Self-hosted MCP gateway: AI agents get audited, SSO-gated, read-only access to PostgreSQL, YugabyteDB, MongoDB and Redis/Dragonfly without ever holding a DB credential. AI-maintained by developerz.ai.
MIT ★ 5 -
@mseep/astra-db-mcpstale
A Model Context Protocol (MCP) server for interacting with Astra DB. MCP extends the capabilities of Large Language Models (LLMs) by allowing them to interact with external systems as agents.
npm Apache-2.0 -
iflow-mcp-azurecosmosdbupdated this year
MCP Server for Azure Cosmos DB - enables LLM agents to interact with Azure Cosmos DB for NoSQL
pypi MIT -
iflow-mcp_azurecosmosdb-azurecosmosdb-mcp-chat-appupdated this year
MCP Server for Azure Cosmos DB - enables LLM agents to interact with Azure Cosmos DB for NoSQL
pypi MIT -
Official Model Context Protocol (MCP) Server for Iceberg DB — Connect Claude Code, Cursor, GitHub Copilot, Gemini & Windsurf to Mumbai-hosted Vector Search & 4-Tier Agent Memory
npm MIT -
sogo-mcp-servergithub.com/shucho/sogo-dbupdated this year
MCP server for Sogo DB — AI agent access to Notion-style .db.json databases
npm MIT -
Agentic AI framework that drafts FDA cybersecurity SRAs for connected medical devices. Vector DB + Graph DB + MCP + LangGraph pipeline, with human-in-the-loop review. Capstone project — reference device: PMx-100.
-
agent_skill_mcp_project (DB MCP + Doc MCP + the agent)
MIT -
Build an MCP server and graphql db to query with an agent
-
MCP server to connect any SQL DB for your agent!
-
Quadraplexgithub.com/neslihaneti/Quadraplexupdated this year
We combine 4 components such as streamlit, mcp, agent and db.
-
mewgithub.com/kash04x/mewupdated this year
CLI MCP tool for AI agents to access redash linked DBs and clickup
-
Memory Kernelgithub.com/artem362/memory-kernelno update data
Local-first memory for AI agents: SQLite FTS5, deterministic recall, no vector DB, no cloud.
pypi -
Persistent, progressive memory for AI agents — encrypted hybrid vector+text knowledge DB with MCP
npm AGPL-3.0 ★ 1 -
mcp-servergithub.com/VVikingsson/mcp-serverupdated this year
MCP Server connecting live DB Schenker Shipment data to your AI agent
MIT -
Driftersgithub.com/mmeents/Driftersupdated this year
Db oriented agent story game thing, lol. Made with ef core for LM Studio characters.
MIT -
Git-native memory for AI agents: append-only, reviewable, no vector DB. Comes with an MCP server.
npm MIT -
Manually-triggered, self-organising personal memory for AI agents — MCP server + CLI. SQLite, no vector DB.
pypi MIT -
memfabricgithub.com/MrBoor/memfabricupdated this year
Self-organizing agent memory. No embeddings, no vector DB.
MIT ★ 1 -
Secure MCP server for safe, read-only DB access by AI agents, with SQL guardrails, table allowlists, PII masking, and audit logs
npm MIT ★ 9 -
nmemgithub.com/6c696e68/nmemupdated this year
Spreading-activation memory for agents. Graph brain, not a vector DB. Rust, MCP, weak-hardware friendly.
MIT -
gogo-mcpgithub.com/donutboyband/gogo-mcpupdated this year
GOGO is a local-first Model Context Protocol (MCP) server that gives AI agents a searchable vector db of your codebase without requiring heavy hardware.
MIT -
面向 Agent 的代码知识图谱:方法级调用图 + 跨服务契约(Feign/HTTP/MQ/DB/Redis),MCP 原生、零原生依赖。
Apache-2.0 ★ 1 -
octoquerygithub.com/benedya/octoqueryupdated this year
Let AI agents query your databases safely: each DB becomes an MCP tool, each schema a skill.
MIT ★ 3 -
Vectorless agent memory. HTML storage + grep retrieval. LongMemEval-S R@5 = 98.9%. No embedding, no vector DB.
MIT -
Local-first personal timeline DB + MCP server — your life, one SQLite file, any agent can read it. 数据不出本机。
MIT ★ 1 -
Agent memory in a single SQLite file — no vector DB, no server, no cloud. Python lib + MCP server + CLI.
MIT -
Read-only PostgreSQL surface for an LLM agent over the MCP protocol, with defense-in-depth at parser, executor, DB-role, and transport layers.
pypi MIT -
nexus-aigithub.com/srikanthanugu892/nexus-aiupdated this year
Enterprise Intelligence Agent — ask your microservice platform anything. AI agent with knowledge graph, semantic search, live DB/API tools, and MCP integration.
MIT -
@gravity-ai/index-mcpactive
MCP server for the Gravity Index — lets AI agents discover developer services (DBs, auth, payments, hosting, etc.) and get install instructions.
npm MIT -
MCP tool servers (hybrid RAG + contract rules DB) driven by a LangGraph agent, with two-layer guardrails, CI-gated retrieval evals and OpenTelemetry tracing. Dockerized, CI-tested, free to run.
MIT -
aionforge-memorygithub.com/jscott3201/aionforge-memoryupdated this year
A Rust-native agentic memory substrate built on the selene-db graph engine: a bi-temporal knowledge graph, hybrid retrieval, and an optional MCP server.
Apache-2.0 ★ 12 -
An MCP server with a built in lightweight proxy server & DB to capture & analyze network traffic within an AI agent
-
context8github.com/hallelx2/context8updated this year
Collective problem-solving memory for coding agents — powered by Actian VectorAI DB. Context7 gives your agent the docs. Context8 gives it what the docs don't cover.
pypi MIT ★ 1 -
AccIntno update data
Local-first MCP Work Model for coding agents. AccInt gives agents two tools, acc_retrieve and acc_act, so work can build memory from real outcomes while the acc.db substrate stays on your machine.
smithery -
dbdensegithub.com/ValkDB/dbdenseupdated this year
Offline database schema context for LLM agents. Extract once, serve via MCP — no live DB needed.
go Apache-2.0 ★ 14 -
Case law for your codebase: is this architecture decision still good law? Clause-level standing graph on Relate (embedded graph DB), MCP preflight for agents, WASM explorer.
Apache-2.0 ★ 1 -
Map a codebase into user journeys (UI → API → code → DB), for you and your AI agent via MCP.
MIT -
safe-migrations-mcpgithub.com/possibly6/safe-migrations-mcpupdated this year
MCP server that stops AI agents from breaking DB schemas and config files. Diff, risk-flag, and require explicit confirmation before any write.
pypi MIT -
synaptic-memorygithub.com/PlateerLab/synaptic-memoryupdated this year
Knowledge graph + MCP tool server for LLM agents with hybrid retrieval, live DB sync, Korean FTS, and memory feedback.
pypi NOASSERTION ★ 33 -
MCP Server to dynamically manipulate by Postgres DB running in docker to use in AI Agents
MIT -
engramgithub.com/petermono56-dev/engramupdated this year
Persistent memory OS for AI agents — typed Markdown vault + a live MCP server (recall/search/write/validate) + operator skill packs. No vector DB, no lock-in. MIT.
MIT -
learn-mcpgithub.com/fvegiard/learn-mcpupdated this year
AI Engineering Learning Reference MCP Server — 550+ searchable references for MCP, RAG, agents, fine-tuning, MLOps, prompt engineering, inference, vector DBs, GPU compute, autonomous agents, and Claude Code workflows
MIT -
Shared memory for AI agents: Claude Code, Codex, Obsidian & GitHub in one searchable SQLite file, served over MCP. Handoff sessions between AIs. No vector DB.
MIT -
dbt-mcp-demogithub.com/gastonlucero/dbt-mcp-demoupdated this year
Analytics stack pairing a read-only PostgreSQL MCP server with dbt and Docker — lets AI agents safely explore the DB and build production-ready dbt models.
Apache-2.0 -
multi-agent-raggithub.com/olegedly/multi-agent-ragupdated this year
Config-driven multi-agent RAG with LangGraph (Researcher → Critic → Synthesizer), pgvector RAG corpus search, and a SolidJS dashboard. Includes a standalone MCP server for IDE integration. LLM-agnostic, hot-reload dev, docker compose DB.
-
Zettelkasten-based persistent memory for AI coding agents. Works with Claude Code, Cursor, VS Code Copilot, Codex, Windsurf & any MCP client. No vector DB — just markdown + git sync.
MIT ★ 144 -
Local-first MCP server that gives AI coding agents durable project memory and token-budgeted context packs. No API keys, no vector DB.
MIT -
ordel-cliupdated this year
Ordel free CLI — local QA-automation for individual devs. Drives the deterministic ordel-engine and exposes it to a BYO coding agent over MCP. No account, no DB, no Ordel LLM.
pypi Apache-2.0 -
engramgithub.com/staticroostermedia-arch/engramupdated this year
Geometric agent memory for Agents: 8-tool lean contract, harness injection at wake, session handoff. 79 MCP tools. Not a vector DB. Rust + MCP.
AGPL-3.0 ★ 21 -
Interactive SQL CLI and MCP server for Metabase. Query any Metabase database from the terminal or from AI agents (Claude, Cursor) without direct DB credentials.
npm -
Local AI memory for coding agents: MCP server with hybrid BM25 + vector search, session capture, stale-memory prune, and git-committable BATTERY.md. SQLite + ONNX embeddings—no cloud DB or embed APIs.
MIT -
gristlegithub.com/Alchemy-Agentic/gristleupdated this year
Graph-based code intelligence for AI agents — traces route→handler→DB end-to-end, scores a change's blast radius, and renders code subgraphs, exposed natively over MCP.
MIT ★ 2 -
mementogithub.com/wmyung/mementoupdated this year
Agent memory: SQLite fact store + Git wiki + keyword bridge. Multi-agent memory system for AI agents. No embeddings, no vector DB. Zero-dependency Python CLI with MCP server.
MIT