What problem are you trying to solve?

Focus on the what and why of the need you have, not the how you'd like it solved.

We want AI assistants (Claude, through the MongoDB Atlas connector) to read our clusters' mongod server logs when investigating issues. The mongodb-logs tool does this, but it's only available in the local MongoDB MCP server. The managed MCP server at https://mcp.mongodb.com doesn't include it, so we can't access server logs there.

What would you like to see happen?

Describe the desired outcome or enhancement.

Add the mongodb-logs tool, or an equivalent log-access tool, to the managed MCP server. It would use the same Atlas account authentication and permissions as the existing tools, so we can read recent server logs for our clusters without running the local server.

Why is this important to you or your team?

Explain how the request adds value or solves a business need.

The local MCP server needs a database connection string, including credentials, stored in plain text in each developer's client configuration (e.g. settings.json). Our security practices don't allow that, so our team doesn't use the local server. The managed server avoids the problem: it authenticates with our Atlas account, stores no database credentials on developer machines, and needs no per-developer setup. Log access is the one capability we need that it doesn't offer.

What steps, if any, are you taking today to manage this problem?

We use the managed MCP server for everything else. For server logs, an engineer leaves the AI workflow, downloads the logs manually from the Atlas UI, and shares the relevant parts. This is slow during incident investigation, and the AI assistant can't correlate logs with the other data it can already reach.