AstroClaw is a cloud-native framework designed to build and deploy AI agents as scalable, secure, and cost-efficient serverless components. It targets developers seeking to decompose monolithic agent processes into manageable, single-responsibility units, leveraging mature cloud ecosystems for operational concerns. The primary benefit is simplified, one-click Infrastructure as Code (IaC) deployment to AWS, resulting in agents that are secure, scalable, and nearly zero-cost when idle.

How It Works

AstroClaw re-architects traditional AI agents by breaking them into independent, single-responsibility serverless functions. This approach leverages cloud-native services for observability, security, and identity management, enhancing robustness and reducing operational overhead. The entire infrastructure is defined as code (IaC), enabling repeatable, single-command deployments. This decomposition facilitates independent scaling and cost optimization, particularly with its default DSQL (serverless, pay-per-request) database option.

Quick Start & Requirements

• Prerequisites: Go 1.22+, Docker (for local PostgreSQL), OpenAI or Anthropic API key.
• Run Locally: Execute go run . after setting OPENAI_API_KEY or ANTHROPIC_API_KEY. A temporary PostgreSQL container starts automatically; data is lost on exit. For persistence, set DATABASE_URL.
• Deploy to AWS: Requires AWS CLI and AWS CDK (npm install -g aws-cdk). Use the interactive ./scripts/deploy.sh script or manual cdk bootstrap and cdk deploy commands within the deploy/aws/infra directory, providing API keys via parameters.
• Post-deployment: Run TUI (go run ./tui/) or CLI (go run main.go) using outputs from the deployment script.
• Database Options: CDK defaults to DSQL (serverless, scale-to-zero) for cost-efficiency. Aurora PostgreSQL is available for full PostgreSQL features, including vector search (pgvector).
• Links: AWS CDK, sqlc

Highlighted Details

• Serverless, single-responsibility component architecture for AI agents.
• One-click IaC deployment to AWS via shell scripts or CDK.
• Choice between cost-effective DSQL and feature-rich Aurora PostgreSQL backends.
• Agents are designed to be secure, scalable, and cost-efficient with near-zero idle costs.

Maintenance & Community

The project is explicitly marked as "early alpha" with an estimated beta release in approximately 8 weeks. Users are encouraged to provide feedback, and the developers acknowledge inspiration from other open-source projects like openclaw and picoclaw. No specific community channels (e.g., Discord, Slack) or dedicated contributor information are provided in the README.

Licensing & Compatibility

The license type is not explicitly stated in the provided README. This omission requires further investigation for commercial use or integration into closed-source projects.

Limitations & Caveats

The project is in an early alpha stage, indicating potential instability and breaking changes. Functionality for third-party applications is planned but not yet detailed. The lack of explicit licensing information is a significant caveat for adoption decisions.