Research data MCP for Trae

Step 1

Trae CN: use the install link

Open this link on the computer where Trae CN is installed. Trae CN shows a confirmation dialog; confirm to add the server.

trae-cn://trae.ai-ide/mcp-import?type=http&name=scholarmuse&config=eyJ1cmwiOiJodHRwczovL3NjaG9sYXJtdXNlLmFpL21jcCJ9

Step 2

Either build: add it manually

Go to Settings > MCP, choose Add > Add Manually and paste this JSON. In the Agent window, set the run environment to local, not cloud.

{
  "mcpServers": {
    "scholarmuse": {
      "url": "https://scholarmuse.ai/mcp"
    }
  }
}

Step 3

Approve in your browser

A scholarmuse.ai window opens. Sign in, approve access and choose which discipline packs Trae may use. There is no API key to create or paste.

Step 4

If no authorization opens

This can happen in the international build, whose documentation does not confirm browser authorization for MCP servers. Use this configuration in its place. It runs mcp-remote on your computer, which opens the sign-in itself, and needs Node.js 18 or later.

{
  "mcpServers": {
    "scholarmuse": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://scholarmuse.ai/mcp"
      ]
    }
  }
}

Step 5

Check that it works

Ask the Trae agent to list the ScholarMuse capabilities. If it cannot see them, check that the server is switched on under Settings > MCP and that the run environment is local.