Feature Request: User Access Control for MCP Servers/Endpoints - Looking to Collaborate

[david-pivonka]

Hi! 👋

I've been exploring MetaMCP and I'm really impressed with the multi-tenancy foundation you've built. I'd love to propose and help implement a feature for user/role-based access control over MCP servers and endpoints.

What I'm thinking

Building on the existing private/public scopes, it would be great to have more granular control over:

• Which users can access specific MCP servers
• Which users can see/use certain namespaces and endpoints
• Maybe even tool-level permissions for sensitive operations

Why I want to help

I have experience with access control systems and I'm excited about MetaMCP's potential in enterprise environments. I think this feature could really unlock wider adoption for organizations that need stricter resource isolation.

Collaboration ask

Rather than diving into implementation, I'd love to discuss:

• What's your vision for access control in MetaMCP?
• Are there any existing patterns or architecture decisions I should follow?
• What would be the preferred approach - database-level, middleware-based, or something else?
• Any concerns or requirements I should consider?

I'm happy to contribute the development work, but want to make sure I'm aligned with your roadmap and technical direction first.

Would you be interested in collaborating on this? Happy to discuss via issues, Discord, or however works best for you!

Thanks for building such a cool project! 🚀

[zhjch05]

Thank you very much. This was a relevant issue about this: #103

I once started it with AI assistance in an old version, but it doesn't really work and become more complex than I thought and I am definitely love to take contributions on this feature.

We can discuss right in this github issue.