openclaw-aws-deploy

What This Skill Does

The openclaw-aws-deploy skill is a specialized infrastructure-as-code automation tool designed to bootstrap a secure, production-ready OpenClaw agent instance on AWS. It automates the lifecycle management of cloud resources, creating a hardened network topology including a private VPC, subnets, and an SSM-ready security group that strictly prohibits inbound ports, ensuring zero SSH surface area. By leveraging ARM64 t4g.medium instances, it optimizes for both cost-efficiency and performance, using AWS SSM to execute configuration scripts securely. The skill manages secret injection via the AWS Parameter Store, keeping your API keys and credentials out of static files. It streamlines the deployment process, from infrastructure creation to the final smoke test, and includes built-in teardown capabilities to prevent cloud sprawl.

Installation

To integrate this into your workflow, install the skill using the ClawKit CLI: clawhub install openclaw/skills/skills/godwinbabu/openclaw-aws-deploy

Ensure your development environment has the aws CLI configured, jq and openssl installed, and your workspace contains a .env.starfish file with your TELEGRAM_BOT_TOKEN and relevant API keys for chosen AI models.

Use Cases

• Cloud Deployment: Rapidly spin up an OpenClaw instance in any AWS region for distributed operation.
• Agent Prototyping: Deploy a fresh EC2 environment to test custom agent behaviors with different LLM backends (Bedrock or Gemini).
• Ephemeral Infrastructure: Use the automated teardown scripts to provision and destroy resources for temporary research tasks, maintaining strict cost control.
• Secure Operations: Transition from local testing to a hardened AWS production environment without exposing administrative ports to the public internet.

Example Prompts

1. "Deploy a new OpenClaw agent named 'research-bot' in us-west-2 using the default configuration."
2. "Update my starfish deployment to use the google/gemini-2.0-flash model and restart the instance."
3. "Tear down the 'starfish' infrastructure using the resources defined in deploy-output.json."

Tips & Limitations

• Security: The skill defaults to SSM-only access. Avoid opening inbound ports manually. Use SSM sessions to interact with your instance.
• Cost Management: Always run the teardown scripts after you finish your experiments to avoid unexpected AWS billing.
• IAM Permissions: Ensure your local AWS credentials have sufficient permissions to create VPC, EC2, IAM, and SSM resources. If you encounter errors, check the logs in deploy-output.json for specific permission denials.