MCP json file over-written with a new nuxt mcp object when multiple Nuxt servers are started in a monorepo setup

[ananthachetan]

Describe the bug

Issue:

I have a monorepo setup that has multiple Nuxt apps. When I start multiple apps each app updates the mcp.json inside Cursor and VSCode folders which a single nuxt MCP object overwriting the previous update done by the other app.

Solution:

If we can have an option to disable the config write that happens when the app starts then we can manually update the mcp json with different nuxt mcp configs (app based)

Current Workaround:

I have manually created different configs in the mcp.json file and whenever another entry gets added when I start an app I delete it:

"mcpServers": { "nuxt-docs": { "url": "https://mcp.nuxt.com/sse" }, "nuxt-sc-base": { "url": "http://localhost:9443/__mcp/sse" }, "nuxt-sc-cdec": { "url": "http://localhost:9444/__mcp/sse" }, "nuxt-sc-cti": { "url": "http://localhost:9445/__mcp/sse" } }

I delete the following config that gets added when the app starts
"nuxt": { "url": "http://localhost:9443/__mcp/sse" }

Reproduction

Start multiple nuxt apps in a monorepo and observe only one entry of "nuxt" mcp is created in the mcp.json with the config of the latest app that was started

System Info

System:
    OS: macOS 15.6.1
    CPU: (10) arm64 Apple M1 Pro
    Memory: 69.38 MB / 16.00 GB
    Shell: 5.9 - /bin/zsh
  Binaries:
    Node: 20.19.5 - ~/.nvm/versions/node/v20.19.5/bin/node
    Yarn: 1.22.5 - ~/.yarn/bin/yarn
    npm: 10.8.2 - ~/.nvm/versions/node/v20.19.5/bin/npm
    pnpm: 10.15.0 - ~/Library/pnpm/pnpm
    bun: 0.1.13 - ~/.bun/bin/bun
  Browsers:
    Chrome: 140.0.7339.134
    Safari: 18.6

Used Package Manager

pnpm

Validations

• Follow our Code of Conduct
• Read the Contributing Guide.
• Check that there isn't already an issue that reports the same bug to avoid creating a duplicate.
• Check that this is a concrete bug. For Q&A, please open a GitHub Discussion instead.
• The provided reproduction is a minimal reproducible of the bug.

Contributions

• I am willing to submit a PR to fix this issue
• I am willing to submit a PR with failing tests (actually just go ahead and do it, thanks!)