Quick Configuration
Choose the setup that matches your environment.
Before You Start
- Confirm you have the account, endpoint, or API key required for AWS.
- Start with minimum scopes and read-only access where possible.
- Keep secrets in environment variables instead of hardcoding them in JSON.
Last verified: 2026-06-13
Environment: uvx with awslabs.aws-api-mcp-server over stdio; uses your AWS profile, read-only mode
Install command:
uvx awslabs.aws-api-mcp-server@latestLocal config
JSON Local config json
{
"aws": {
"command": "uvx",
"args": ["awslabs.aws-api-mcp-server@latest"],
"env": {
"AWS_REGION": "us-east-1",
"AWS_API_MCP_PROFILE_NAME": "${AWS_PROFILE}",
"READ_OPERATIONS_ONLY": "true"
}
}
} After copying
Check the AWS JSON before running it in OpenClaw, then verify the server with a real prompt.
Verify prompt
List my S3 buckets and the EC2 instances in us-east-1.
Expected success output
You should see your buckets and instances returned via read-only AWS API calls.
Common Pitfalls & Fixes
- ⚠️ Watch out: IAM permissions, region selection, and resource naming.
- 🔑 Always store API keys in environment variables, never hardcode them in JSON.
- 🛡️ Start with read-only scopes if available to verify connection safely.
Security checklist
- Keep READ_OPERATIONS_ONLY=true and use a least-privilege IAM profile.
- Never inline AWS keys in JSON; use an AWS profile or environment credentials.
Common errors
| Error | Cause | Fix |
|---|---|---|
uvx not found | uv is not installed or not on PATH. | Install uv (which provides uvx). |
Unable to locate credentials | No AWS profile or credentials are configured. | Configure an AWS profile and set AWS_API_MCP_PROFILE_NAME to it. |
AccessDenied | The IAM identity lacks the needed read permissions. | Attach least-privilege read-only policies for the services you query. |
Example Prompts
Once connected, try these prompts to test capabilities:
- List S3 buckets and their total size.
- Check Lambda function runtime and memory settings.
- Get EC2 instance status and public IP addresses.
Verification Checklist
- Run 1-2 real prompts to confirm AWS returns usable data.
- Check that error messages are clear enough for troubleshooting.
- Document the required scopes, dependencies, and env vars for future reuse.