Skip to main content

Amazon Q Developer CLI Integration with Harness MCP Server

What is Amazon Q Developer​

Amazon Q Developer is a generative AI-powered assistant by AWS built on Amazon Bedrock. It helps with understanding, building, extending, and operating AWS applications. You can use it in various environments including the CLI (command line), IDEs, AWS Console, chat applications, etc. It supports features like conversational assistance, code suggestions, inline completions, and translating natural language into shell commands. 

In the CLI context, Amazon Q Developer enhances your terminal experience by: • providing command autocompletion for many common CLIs (like git, npm, docker, aws)  • enabling “chat mode” where you can ask questions in natural language from the terminal and get responses that help you build, debug, or operate your code or infrastructure  • translating plain language instructions into executable shell commands  • understanding context (files in the project, existing settings, AWS resource state) so that suggestions and completions are more relevant 

There are pricing tiers including a Free Tier; Amazon Q Developer provides features in its Free and Pro tiers. 

Prerequisites​

  • Amazon Q Developer CLI installed and configured with AWS credentials
  • Harness account with an API Key (Org ID and Project ID optional)
  • Go installed to build the Harness MCP Server
  • Docker Installed to run the Harness MCP

Step 1: Build the Harness MCP Server​

Clone the MCP server repository

git clone https://github.com/harness/mcp-server.git
cd mcp-server

Build the binary if building locally

go build -o cmd/harness-mcp-server/harness-mcp-server ./cmd/harness-mcp-server

# Make executable if needed
chmod +x harness-mcp-server

⸻

Step 2: Run the Harness MCP Server​

HARNESS_API_KEY=<YOUR_API_KEY> \
HARNESS_DEFAULT_ORG_ID=<YOUR_ORG_ID> \
HARNESS_DEFAULT_PROJECT_ID=<YOUR_PROJECT_ID> \
./cmd/harness-mcp-server/harness-mcp-server stdio

You can also set HARNESS_BASE_URL if using a custom Harness domain.

Step 3: Configure Amazon Q Developer CLI​

Edit or create the MCP configuration file at: ~/.aws/amazonq/mcp.json

{
  "mcpServers": {
    "harness": {
      "command": "/path/to/harness-mcp-server",
      "args": ["stdio"],
      "env": {
        "HARNESS_API_KEY": "<YOUR_API_KEY>",
        "HARNESS_DEFAULT_ORG_ID": "<YOUR_ORG_ID>",
        "HARNESS_DEFAULT_PROJECT_ID": "<YOUR_PROJECT_ID>",
        "HARNESS_BASE_URL": "<YOUR_BASE_URL>"
      }
    }
  }
}

Step 4: Query Harness with Amazon Q​

Start a session with:

q chat

Then interact with Harness using natural language prompts.

Sample Prompts​

  • List all my pipelines in Harness
  • What repositories are linked to my account?
  • Show me all artifacts in my project
  • Find feature flags created in my org
  • Summarize the last 5 deployments in project XYZ
  • Which connectors are configured in Harness?

Troubleshooting​

  • If Amazon Q cannot connect to Harness MCP, check the command path, arguments, and environment variables in mcp.json.
  • If you see credential errors, regenerate the API key in Harness.
  • If no resources are returned, confirm that Org ID, Project ID, or Base URL are set correctly.
  • For performance or latency issues, ensure the MCP server has sufficient resources and review its logs.