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.
-
Multi-tenant Google Slides MCP server for AI clients. Create and manage presentations via MCP. Node.js + Cloud Run.
MIT -
google-docs-mcp-servergithub.com/prixieroxanbooc/google-docs-mcp-serverupdated this year
Multi-tenant Google Docs MCP server for AI clients. Create, read, edit documents via MCP. Node.js + Cloud Run.
MIT -
Remote Google Health (Fitbit/Pixel Watch) MCP server with built-in OAuth 2.0, ready for Google Cloud Run's free tier.
MIT -
raceai-gcpgithub.com/kamdinshafiepour/raceai-gcpupdated this year
RaceAI rebuilt on Google Cloud — Gemini API, MCP server, FastAPI, Docker, Cloud Run
-
gsheets-mcp-servergithub.com/prixieroxanbooc/gsheets-mcp-serverupdated this year
Multi-tenant Google Sheets MCP server for AI clients. Read, write, manage spreadsheets via MCP. Node.js + Cloud Run.
MIT -
gmail-mcp-servergithub.com/prixieroxanbooc/gmail-mcp-serverupdated this year
Multi-tenant Gmail MCP server for AI clients (Claude, ChatGPT). Built with Node.js, deployed on Google Cloud Run.
MIT -
Copilot-MCPgithub.com/pedroalexleite/Copilot-MCPupdated this year
MCP server exposing real-time financial data tools to Microsoft Copilot Studio via Google Cloud Run for a Copilot's Studio agent published in Microsoft Teams.
-
MCP-native agent graphs with a real UI. Local-first / BYOK — TypeScript monorepo (Next.js + Electron + MCP server + LangGraph). Powered by Google Gemini + Vertex AI on Cloud Run. Submitted to the Google for Startups AI Agents Challenge.
MIT -
"AI agent that answers location questions using 7 geospatial tools (site scoring, flood risk, LiDAR, street-view walkability, land cover), with a 3D CesiumJS twin. Runs live on Google Cloud Run."
MIT -
Real-time MBTA transit data (Boston) exposed through a Model Context Protocol (MCP) server, built with Python and Streamable HTTP and deployed on Google Cloud Run.
-
Even your AI needs a second opinion. A self-hosted MCP server on Google Cloud Run that lets Claude, ChatGPT, Cursor, and Gemini CLI peer-review their work with a different frontier model — Claude Opus 5.5 or Gemini 3.8 Flash on Vertex AI — with PII scrubbed before anything leaves your server.
Apache-2.0 -
ctxHub: version-controlled context for AI agents. Commit your project's decisions from any chat, review every change, and pull it into Claude, ChatGPT or the terminal, without losing what changed or why. MCP server on Google Cloud Run.
-
Persistent Project Context for Google Gemini — IANA-registered .faf format · MCP server + Cloud Run REST API
pypi MIT ★ 2 -
Google Analytics 4 MCP server — local STDIO and team Cloud Run modes
MIT -
This project implements a weather data service using the Model Context Protocol (MCP), designed to be consumed by intelligent agents (e.g., LLM agents). It's built with Python, containerized with Docker, and deployed to Google Cloud Run with continuous deployment configured from GitHub.
★ 1 -
STDIO MCP server that allows agents to run their code in Google Cloud Functions
-
gmail-skillgithub.com/hec-ovi/gmail-skillupdated this year
Gmail for AI agents: OAuth login, search, read threads, draft, send with attachments, labels. Five Agent Skills + a verb CLI + a FastMCP stdio server over the Gmail API. Needs your own free Google Cloud project. Untrusted-mail fencing, dry-run sends, contract-isolated Python 3.11+ layers.
MIT ★ 1 -
MCP server to deploy apps to Cloud Run
go Apache-2.0 ★ 639 -
Evidence-first Medicare Stars agent on GCP, Vertex AI, ADK, MCP, and BigQuery, privately deployed to Cloud Run with fail-closed guardrails, CVE-gated OCI CI/CD, evaluations, and per-request cost telemetry.
NOASSERTION -
GenSpark-focused remote Streamable HTTP MCP server for Google Drive/Docs/Sheets/Slides
MIT ★ 2 -
Create-only Google Workspace admin MCP server for employee onboarding — availability checks, account creation, and credentials delivery with Google OAuth and keyless domain-wide delegation.
MIT -
—
Apache-2.0 -
Portfolio of Li ZhenXi (SRE, cloud and AI infrastructure), run as a production service: Go on Cloud Run with its own live SLO, an MCP server for AI agents, OpenTelemetry traces, DORA metrics and build-once keyless deploys; Next.js and Terraform
MIT ★ 1 -
identity-verifier-mcpgithub.com/prixieroxanbooc/identity-verifier-mcpupdated this year
AI-powered identity verification MCP server with face matching, liveness detection, and OCR. Python + Cloud Run.
MIT -
hubspot-mcp-servergithub.com/prixieroxanbooc/hubspot-mcp-serverupdated this year
Multi-tenant HubSpot CRM MCP server for AI clients. Manage contacts, companies, deals via MCP. Python + Cloud Run.
MIT -
An MCP-native agentic control plane that installs into your own Google Cloud project in one command. Serverless git over Firestore and GCS, post-quantum encryption at rest, and a signed, journalled audit trail behind every privileged action.
Apache-2.0 ★ 1 -
Servidor MCP en Python para registrar y consultar precios de supermercado en Google Sheets
MIT -
Deployed retrieval-augmented agent: ADK + authored MCP server + pgvector on Cloud Run, with an eval harness
MIT -
Autonomous line producer for film shoots — Gemini + CP-SAT + Grafana MCP. The LLM formulates, the solver decides, a human approves.
Apache-2.0 -
GLOBAL-email-RAG-MCPgithub.com/eqramul01/GLOBAL-email-RAG-MCPupdated this year
v2: Hybrid (BM25+vector) RAG over a six-figure legal email archive — full-text message search, recipient-graph filtering, in-chat .eml reading. Gemini Embedding 2, LanceDB, Cloud Run, MCP, queryable from Claude.ai with cited source links.
MIT -
rag-X-mcp-v1github.com/Sarthak6o1/rag-X-mcp-v1updated this year
MCP-ready RAG monorepo: FastAPI hybrid search, FastMCP tool server, admin upload portal.
Apache-2.0 -
AskHooman: asks and tasks between people and their AI agents. An MCP server + app API, data in each user's own Google Drive
Apache-2.0 -
Shadow-AI discovery and EU AI Act / SOC 2 compliance auditing for GCP — ADK agent, custom MCP server, deterministic checks, hashed evidence, human-gated remediation.
-
bq-data-agentgithub.com/farjanaferdausi-cs50ai/bq-data-agentupdated this year
A reasoning AI agent that investigates BigQuery data and answers business questions with evidence, not guesses — built with Gemini, ADK & BigQuery MCP Server, deployed on Cloud Run.
-
Negotiate with an AI car dealer that must keep its floor price secret. Three LLM-security levels (ADK, LangGraph, MCP, A2A) measured with evals, deployed on Cloud Run.
★ 1 -
Read-only BigQuery MCP server for Claude, Cursor, Windsurf, Copilot and other MCP clients. Auto-LIMIT, dry-run cost guard, Application Default Credentials. Run with uvx.
pypi MIT ★ 1 -
An autonomous data-ops analyst on Google ADK that predicts a KPI miss, finds the cause, and opens the pull request that fixes it.
MIT -
workspace-agentgithub.com/rashidalmarri-dev/workspace-agentupdated this year
Enterprise Google Chat assistant that reads Workspace context (Gmail, Drive, Docs) and writes deliverables to a Shared Drive, split across two MCP servers so read-only personal data and read/write actions never share credentials. Built on Google ADK + Vertex AI Gemini with slash-command workflows and runtime access control.
MIT -
A WebMCP-native live auction house where AI agents bid inside human-set mandates they structurally cannot exceed. Your agent bids. You set the limits. Everything on the record.
MIT -
MCP server + multi-agent orchestrator over FRED economic data: constrained tool contracts, a supervisor delegating to specialist agents, prompt-injection containment, a token-budget guardrail, and a CI-gated evaluation harness. Runs fully offline with no API key.
★ 1 -
Describe your app once. Deploy it anywhere: one spec for Cloudflare, your own VPS, Google Cloud, AWS, Kubernetes and more, with plan before apply and no state file.
go Apache-2.0 ★ 1 -
Protection-first BigQuery MCP server — every query is dry-run and refused unless BigQuery itself classifies it as a single SELECT inside the dataset allowlist and the byte budget, then runs as a named job under maximumBytesBilled. Rows come back column-keyed with explicit caps; failures carry a retry hint
MIT