AI Coding Assistant · AWS / Amazon

Give Amazon Q Developer Custom Browser-Based MCP Tools

Amazon Q Developer added MCP support in 2025, letting its VS Code and JetBrains extensions call external tools. Connect Agent MCP Studio's relay bridge to give Q's agent access to browser-built Python scripts, SQL queries, and API integrations — no local server required.

Share:

Prerequisites

Step-by-Step Setup

  1. Get your relay URL and download bridge.js

    Open the studio → SettingsMCP Relay Bridge → copy URL → click Connect. Then:

    curl -O https://agentmcp.studio/bridge.js && npm install ws
  2. Create the Amazon Q MCP config

    Amazon Q Developer reads MCP servers from:

    • Global: ~/.aws/amazonq/mcp.json
    • Project: .amazonq/mcp.json in your workspace root
    {
      "mcpServers": {
        "agent-mcp-studio": {
          "command": "node",
          "args": [
            "/path/to/bridge.js",
            "wss://agentmcp.studio/api/relay/YOUR-UUID"
          ],
          "timeout": 30000
        }
      }
    }
  3. Enable Agent mode in Amazon Q

    Open the Amazon Q chat panel in VS Code. Switch to Agent mode using the mode selector at the bottom of the chat. The MCP tools will be automatically loaded — you'll see a tools indicator in the interface.

  4. Use tools in Q's agent

    Ask Q to use one of your tools: "Use the query_database tool to show me the most recent 10 orders from our database." Q will call bridge.js, relay the request to your browser, and return the result inline.

AWS context + MCP tools

Amazon Q has native AWS context (IAM, CloudFormation, Lambda). Combining Q's AWS knowledge with your custom MCP tools is powerful — for example, a tool that queries your DynamoDB or reads S3 data, with Q already knowing your AWS architecture.

Frequently Asked Questions

Yes — Amazon Q Developer CLI (v1.7+) and the VS Code extension support MCP servers. Configure servers in your Q Developer config or extension settings and Q's agent can call Agent MCP Studio tools during coding sessions.

For Q Developer CLI: edit ~/.aws/amazonq/mcp.json. For the VS Code extension: use the Q Developer extension settings under "MCP Servers". Both use the same server definition format as other MCP clients.

Amazon Q Developer already has native AWS service awareness. Adding Agent MCP Studio via MCP extends Q with additional custom tools — Python functions, SQL, external APIs — that complement Q's built-in AWS knowledge.

Amazon Q Developer has a free tier available without an AWS account, but full agentic features including MCP tool calling may require an Amazon Q Developer Pro subscription or AWS Builder ID login.

Check: (1) Q Developer CLI is version 1.7 or newer, (2) mcp.json is in the correct location and valid JSON, (3) Node.js is in PATH, (4) relay UUID matches Studio Settings, (5) Agent MCP Studio tab is open and Connected.

Related Integrations