Vee3

Connect Vee3 MCP to VS Code

Published: July 8, 2026

Ask your agent

Recommended

Copy and paste this into VS Code Copilot chat (agent mode) and let the agent walk you through setup.

Agent Instruction

Connect to VS Code

Click Add to VS Code. VS Code opens an MCP install flow. Confirm your API key, pick workspace or user scope, then install. You may need to restart VS Code before tools appear. Use Copilot agent mode so tools can run.

In VS Code's install dialog

  1. Name: defaults to vee3; you can rename it
  2. Headers: confirm VEE3_API_KEY is your API key
  3. Choose workspace or user scope, then confirm install
  4. Restart VS Code if tools do not appear right away

Tip: set VEE3_API_KEY in your environment before launching VS Code, then leave the header as ${env:VEE3_API_KEY} in the install config.

Add to VS Code
Manual setup via .vscode/mcp.jsonShow
  1. In your project, create .vscode/ if needed, then add .vscode/mcp.json. Or run MCP: Open User Configuration for all workspaces.
  2. Paste the config below (merge the vee3 entry if you already have other servers).
  3. Start the server from the editor (code lenses on the file) or MCP: List Servers. Restart VS Code if the server or tools do not show up.

.vscode/mcp.json

Workspace-scoped; safe to commit if the header uses an env variable, not a literal key.

{
  "servers": {
    "vee3": {
      "type": "http",
      "url": "https://mcp.vee3.io/mcp",
      "headers": {
        "VEE3_API_KEY": "${env:VEE3_API_KEY}"
      }
    }
  }
}

Verify the connection

In Copilot agent mode, ask the agent to call meta-tools.list_groups or search the catalog with meta-tools.search. If tools return data, the connection is working.

Customize which tools appear

From the dashboard, open Customize MCP to turn capability groups on or off. Changes apply to your MCP server on the next connection.

Next steps