Get started

Create the harness with a name and execution role:

aws bedrock-agentcore-control create-harness \
  --harness-name "MyHarness" \
  --execution-role-arn "arn:aws:iam::123456789012:role/MyHarnessRole"

Poll get-harness until "status": "READY". Note the arn in the response.

aws bedrock-agentcore-control get-harness \
  --harness-id "MyHarness-XyZ123"

Invoke from Python. If you don’t specify a model, the harness defaults to Anthropic Claude Sonnet 4.6 on Amazon Bedrock:

import boto3

client = boto3.client("bedrock-agentcore", region_name="us-west-2")

response = client.invoke_harness(
    harnessArn="arn:aws:bedrock-agentcore:us-west-2:123456789012:harness/MyHarness-XyZ123",  # Replace with your harness ARN
    runtimeSessionId="1234abcd-12ab-34cd-56ef-1234567890ab",
    messages=[{
        "role": "user",
        "content": [{"text": "Research three tropical vacation options under $3k."}]
    }],
)

for event in response["stream"]:
    if "contentBlockDelta" in event:
        delta = event["contentBlockDelta"].get("delta", {})
        if "text" in delta:
            print(delta["text"], end="", flush=True)
    elif "runtimeClientError" in event:
        print(f"\nError: {event['runtimeClientError']['message']}")

Install the AgentCore CLI:

npm install -g @aws/agentcore

Create a harness project non-interactively with flags:

agentcore create --name myresearchagent --model-provider bedrock

Deploy and invoke:

agentcore deploy
agentcore invoke --harness myresearchagent \
  --session-id "$(uuidgen)" \
  "Research three tropical vacation options under $3k, within five hours of NYC."

The response streams to your terminal. Reuse the same --session-id across invocations to continue the conversation in the same environment.

To add more harnesses to an existing project, use agentcore add harness. To generate a standalone Python invoke script, add --with-invoke-script.

Useful flags:

Install the AgentCore CLI:

npm install -g @aws/agentcore

Run agentcore create without flags to launch the interactive wizard:

agentcore create

After confirmation, the wizard scaffolds your project. Deploy with agentcore deploy, then invoke with agentcore invoke.

Check project status at any time with agentcore status:

If you want to test the harness in your local environment, you can run the dev server:

agentcore dev

When you run agentcore dev, the CLI first deploys your harness resources to AWS, creating the IAM role, harness, and any memory or credentials configured in your project:

Once deployment completes, it starts a local server and opens the agent inspector in your browser where you can chat with the harness, inspect traces, and browse project resources:

Expand Harness Settings to view and override the harness configuration for the current session: