The PromptRCA Project

An open source, multi-agent AI system for automated root cause analysis of AWS serverless applications. Built by the community, for the community.

Why Open Source?

We believe that powerful debugging tools should be accessible, transparent, and community-driven

Community-Driven

Built by AWS engineers, for AWS engineers. Your contributions shape the future of the project.

Transparency

Full source code access. Understand exactly how investigations work and verify the results.

Trust & Security

Open source means security through transparency. Community reviews and contributions make it stronger.

Two Editions, One Mission

Choose the edition that fits your needs. Both share the same powerful core technology.

Community Edition

Open Source

The core RCA agent logic, fully open source and available for everyone. Deploy it yourself or use it as a Bedrock agent.

  • License: AGPL-3.0 (Open Source)
  • Deployment: Lambda, Server, ECS, Development
  • Standalone: Use independently as Bedrock agent
  • Core Features: Multi-agent system, 50+ AWS tools
View on GitHub
Enterprise

Enterprise Edition

Full-stack application with web interface, user management, and complete AWS infrastructure. Built for teams and organizations.

  • License: AGPL-3.0 with commercial restrictions
  • Features: Web UI, user management, API Gateway
  • Infrastructure: Complete Terraform deployment
  • Integration: Uses Community Edition as submodule
Try Enterprise Edition

Technical Architecture

A multi-agent system designed for comprehensive AWS incident investigation

1

Lead Analyst

Intelligent coordinator that orchestrates specialist agents and synthesizes findings

10+

Specialist Agents

Service-specific agents for Lambda, API Gateway, Step Functions, IAM, and more

50+

AWS Tools

Comprehensive toolset for investigating AWS services and configurations

12+

AWS Services

Coverage across Lambda, API Gateway, Step Functions, IAM, DynamoDB, S3, and more

Key Components

Core Services

  • • Lambda (9 tools)
  • • API Gateway (8 tools)
  • • Step Functions (8 tools)
  • • IAM (7 tools)
  • • X-Ray & CloudWatch

Additional Services

  • • DynamoDB, S3, SQS, SNS
  • • EventBridge
  • • VPC/Network
  • • Meta-tooling capabilities
  • • Dynamic tool creation

How to Contribute

We welcome contributions from the community! Every contribution, big or small, makes a difference.

Code Contributions

Submit pull requests for bug fixes, new features, or improvements. We follow standard GitHub workflows.

View Repository →

Documentation

Help improve our docs, write tutorials, or fix typos. Great documentation helps everyone.

Contribute Docs →

Bug Reports

Found a bug? Report it on GitHub Issues. Include steps to reproduce and environment details.

Report Bug →

Feature Requests

Have an idea? Share it! We're always looking for ways to improve PromptRCA.

Suggest Feature →

Project Vision

Our goal is to make AWS incident analysis accessible, intelligent, and community-driven

Current State

  • Multi-agent architecture with 10+ specialist agents
  • 50+ AWS investigation tools across 12+ services
  • Open source Community Edition (AGPL-3.0)
  • Enterprise Edition with full-stack UI

Future Goals

  • Expand AWS service coverage
  • Enhanced multi-agent coordination
  • Community-driven tool development
  • Improved documentation and tutorials

Join the Community

Help shape the future of AWS incident analysis. Your contributions, feedback, and ideas are welcome!

Resources & Links

Everything you need to get started with PromptRCA

GitHub Repository

Browse the source code, submit issues, and contribute to the project.

Visit Repository →

Documentation

Learn how to deploy, configure, and use PromptRCA in your environment.

Read Docs →

License

PromptRCA Community Edition is licensed under AGPL-3.0. Free to use, modify, and distribute.

View License →

Contact

Have questions or want to get in touch? We'd love to hear from you.

Send Email →

Ready to Get Involved?

Join the open source community and help build the future of AWS incident analysis