Three Stars ⭐⭐⭐

The fastest AI agent deployment tool for prototyping on AWS.

Three Stars doesn't require you to set up CDK, wait for CloudFormation stack deployments, or struggle with circular dependency errors. All you need is Python — the same language you use to develop your AI agent. Our promise and principles are as follows.

• Speed — Deployment speed is our top priority
• Steps — Streamlined steps including tool setup and error handling
• Small — Just enough for prototyping

Three Stars provides the sss command (short for speed step small) for quick deployment. Commands are available both as a CLI and as an MCP server for your AI coding agent.

Quick Start

Prerequisites:

• Python 3.12+
• AWS credentials configured (aws configure)
• Permissions for S3, CloudFront, IAM, Lambda, and Bedrock AgentCore

Use with AI Agents (Recommended)

The easiest way to use three-stars is through an AI coding agent. three-stars provides scaffolding tools for agents like Claude Code.

Add this to your Claude Code MCP settings.

{
  "mcpServers": {
    "three-stars": {
      "command": "uvx",
      "args": ["--from", "three-stars", "three-stars-mcp"]
    }
  }
}

Install as CLI

pip install three-stars

This installs the sss command. From zero to deployed in four lines:

pip install three-stars
sss init my-app && cd my-app
sss deploy
# Open the printed CloudFront URL in your browser

How Commands Work

Scaffold a Project : sss init

sss init my-app
cd my-app

This creates the following structure:

my-app/
├── three-stars.yml     # Configuration
├── app/                # Frontend — use any framework (React, Vue, etc.)
│   └── index.html
└── agent/              # AI agent (Python)
    ├── agent.py        # Strands Agent with SSE streaming
    ├── tools.py        # MCP tool loader
    ├── memory.py       # AgentCore Memory session manager
    └── mcp.json        # MCP server configuration

The app/ directory starts with a plain HTML file, but you can replace it with your favorite frontend framework — React, Vue, Svelte, or anything that builds to static files. Just point app.source in the config to your build output directory.

The agent/ directory contains a starter Strands Agent that streams responses as Server-Sent Events. It supports:

• MCP Tools — add tool servers in agent/mcp.json (stdio and HTTP transports). Environment variable references (${VAR}) and AWS credentials are forwarded automatically.
• Conversation Memory — when AgentCore Memory is configured, conversation history is preserved across turns within a session.

three-stars.yml controls your deployment:

name: my-ai-app
region: us-east-1

agent:
  source: ./agent
  model: us.anthropic.claude-sonnet-4-6  # Any Bedrock model ID
  description: "My AI assistant"

app:
  source: ./app
  index: index.html

api:
  prefix: /api

Deploy to AWS : sss deploy

Deploy your application to AWS.

sss deploy

First deploy typically completes in ~5 minutes.

[1/5] S3 storage ready                   0:00:01
[2/5] AgentCore ready                    0:00:48
[3/5] Lambda@Edge function ready         0:00:04
[4/5] CloudFront distribution deployed   0:00:45
[5/5] AgentCore resource policy set      0:00:02

     Post-Deployment Health Check
┌────────────┬───────────────────┬──────────┐
│ Resource   │ ID / Name         │ Status   │
├────────────┼───────────────────┼──────────┤
│ S3 Bucket  │ sss-my-app-…      │ Active   │
│ AgentCore  │ rt-abc123         │ Ready    │
│ CloudFront │ E1234567890       │ Deployed │
└────────────┴───────────────────┴──────────┘

Deployed successfully!
URL: https://d1234567890.cloudfront.net

Open the URL to see your AI agent chat app — the frontend streams responses in real time with Markdown rendering and tool call indicators.

graph LR
    Browser["User Browser"]
    CF["CloudFront Distribution<br/>(HTTPS CDN)"]
    S3["S3 Bucket<br/>(Static Frontend)"]
    Edge["Lambda@Edge<br/>(SigV4 Signing)"]
    AC["Bedrock AgentCore<br/>(AI Agent Runtime)"]

    Browser --> CF
    CF -- "/*" --> S3
    CF -- "/api/*" --> Edge --> AC

Subsequent deploys are even faster because dependencies are cached and only changed resources update.

sss deploy  # ~23 seconds on redeploy

Check Status : sss status

Show deployment status of all AWS resources.

sss status

Use --sync to discover actual resources from AWS and update the local state file:

sss status --sync

Tear Down : sss destroy

Remove all deployed AWS resources.

sss destroy

Use --name to discover and destroy resources by project name when the state file is missing:

sss destroy --name my-app --region us-east-1

Note: Lambda@Edge functions cannot be deleted immediately. AWS cleans up edge replicas asynchronously after the CloudFront distribution is removed, which can take 30–60 minutes. If replicas still exist, sss destroy will report that the function remains and you can safely re-run the command later to finish cleanup.

Development

To use the MCP server from a local checkout, point to the source directory:

{
  "mcpServers": {
    "three-stars": {
      "command": "uv",
      "args": ["--directory", "/path/to/three-stars", "run", "three-stars-mcp"]
    }
  }
}

# Install in development mode
uv sync

# Run tests
uv run pytest

# Lint
uv run ruff check three_stars/ tests/

# Format
uv run ruff format three_stars/ tests/

License

Apache License 2.0