MyMCPs
Enables routing to Notion's remote MCP server using OAuth, with automatic discovery of OAuth metadata and token storage.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MyMCPsshow me all tools from my upstream MCPs"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MyMCPs
MyMCPs is a self-hosted Model Context Protocol (MCP) gateway. Connect your AI client to one endpoint, then manage every upstream MCP, credential, and access token from one dashboard.
What it does
Connects HTTP and npm-based MCP servers.
Supports bearer tokens, custom headers, and OAuth.
Issues access tokens for all MCPs or a selected set.
Exposes every allowed upstream through
GETandPOST /mcp.Records gateway activity and usage analytics.
MyMCPs is self-hosted and invite-only. The first user becomes the administrator during onboarding.
Related MCP server: Shared MCP Gateway
Run locally
You need Node.js 24 or newer, pnpm, and Deno if you want to run npm-based MCPs.
pnpm install
cp .env.example .env
node ace generate:key
node ace migration:run
pnpm run devOpen http://localhost:3333 and create the admin account.
Connect an AI client
Add your upstream servers from MCPs.
Create a token from Access tokens.
Point your AI client to
https://your-domain.example/mcp.Send the token as
Authorization: Bearer <token>.
Tools use {mcp-slug}__{tool-name} names by default. For clients with large tool catalogs, add the following request header to enable progressive discovery:
X-MyMCPs-Tool-Mode: lazyLazy mode exposes list_mcps, tool_search, and call_tool instead of loading every tool definition at once.
OAuth MCPs
For providers that support MCP OAuth discovery and dynamic client registration, choose OAuth when adding the server, save it, then select Connect OAuth. Set APP_URL to the instance's public HTTPS URL so callback URLs are generated correctly.
How to deploy to my Coolify
This repository includes a production Docker image, a Compose service, and a coolify.json profile.
In Coolify, create a project and add a Public Repository resource.
Paste
https://github.com/0xtlt/MyMCPsas the repository URL and select the branch you want to deploy.Use Docker Compose as the build pack and
/docker-compose.ymlas the Compose file. Coolify may fill these settings fromcoolify.json.Add a domain to the
mymcpsservice and setAPP_URLto the same HTTPS origin, for examplehttps://mcp.example.com.Deploy, open the domain, and complete onboarding.
The deployment exposes port 3333, checks /health, and runs database migrations before the app starts. The mymcps-data volume persists SQLite, encrypted secrets, the generated app key, and Deno sandbox data under /app/tmp.
APP_KEY is required, but you do not need to create it in Coolify. On the first start, the container generates a valid key, saves it to /app/tmp/app.key, and reuses it on every deploy. Back up the mymcps-data volume and do not rotate the key, or existing encrypted MCP credentials will become unreadable.
The Coolify profile sets TRUST_PROXY=true so logs use the forwarded client IP instead of the proxy's IP.
Useful commands
pnpm run dev # Start the development server
pnpm test # Run all tests
pnpm run lint # Check code style
pnpm run typecheck # Check TypeScript
pnpm run build # Create a production buildStack
MyMCPs uses AdonisJS 7, Inertia, React 19, SQLite, the MCP TypeScript SDK, and Deno. See SECURITY.md for the deployment security baseline and vulnerability reporting process.
License
MIT © 2026 Thomas Tastet.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- FlicenseCqualityDmaintenanceA powerful gateway for the Model Context Protocol (MCP) that unifies AI toolchains by federating multiple MCP servers, wrapping REST APIs as MCP tools, and supporting multiple transport methods with an admin dashboard.Last updated1
- Flicense-qualityBmaintenanceAggregates multiple shared MCP servers into a single HTTP gateway, providing a unified, observable, and reusable MCP access layer for multiple AI clients like Codex, OpenCode, and OpenClaw. It centralizes configuration management, logging, health checks, and circuit breaking to simplify multi-client MCP deployments.Last updated5
- Flicense-qualityDmaintenanceUnified local MCP AI Gateway that routes across Groq, OpenRouter, Mistral, and local Ollama providers, with OpenAI-compatible APIs, MCP tools, fallback/racing router, monitoring, and web dashboard.Last updated
- Alicense-qualityCmaintenanceSelf-hosted MCP gateway that connects productivity tools like Jira, Confluence, and Obsidian to AI assistants via a single endpoint.Last updated2MIT
Related MCP Connectors
Self-hosted MCP gateway: turn any API, database or MCP server into AI connectors — no code.
MCP server for AI agents to plan, verify, and deploy Cloudflare-native apps.
Self-hosted federated MCP gateway: one OAuth 2.1 MCP server in front of N apps, user-level scopes.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/0xtlt/MyMCPs'
If you have feedback or need assistance with the MCP directory API, please join our Discord server