-
@@ -20,7 +20,7 @@ >[!IMPORTANT] ->This repository leverages [SourceSage](https://github.com/Sunwood-ai-labs/SourceSage), and approximately 90% of the release notes, README, and commit messages are generated using [SourceSage](https://github.com/Sunwood-ai-labs/SourceSage) + [claude.ai](https://claude.ai/). +>This repository leverages [SourceSage](https://github.com/Sunwood-ai-labs/SourceSage). Approximately 90% of the release notes, README, and commit messages are generated using [SourceSage](https://github.com/Sunwood-ai-labs/SourceSage) and [claude.ai](https://claude.ai/). >[!NOTE] >AMATERASU is the successor project to [MOA](https://github.com/Sunwood-ai-labs/MOA). It has evolved to run each AI service on an independent EC2 instance using Docker Compose, enabling easy deployment with Terraform. @@ -28,44 +28,44 @@ ## 🔒 Security-Focused Design Philosophy -AMATERASU is a private AI platform specifically developed for Japanese enterprises with stringent security requirements. It enables the secure use of LLMs based on AWS Bedrock: +AMATERASU is a private AI platform foundation developed specifically for Japanese enterprises with stringent security requirements. It enables the secure use of LLMs based on AWS Bedrock: - **Secure LLM Foundation with AWS Bedrock**: - - Supports the Claude-3 model optimized for enterprises - - AWS's enterprise-grade security - - Granular access control based on IAM roles + - Supports the Claude-3 model, optimized for enterprise use. + - Leverages AWS's enterprise-grade security. + - Fine-grained access control using IAM roles. - **Operation in a Completely Closed Environment**: - - Operates only within the internal network - - Supports private cloud/on-premises deployments + - Operates only within the internal network. + - Supports private cloud/on-premises deployments. - **Enterprise-Grade Security**: - - Access control via IP whitelist - - HTTPS/TLS encrypted communication - - Network segmentation using AWS Security Groups - - IAM role management based on the principle of least privilege + - Access control via IP whitelisting. + - HTTPS/TLS encrypted communication. + - Network segmentation using AWS Security Groups. + - IAM role management based on the principle of least privilege. -## 🏢 Key Features +## ✨ Key Features ### 1. Secure ChatGPT-like Interface (Open WebUI) -- Provides an internal chat UI -- Prompt template management -- Conversation history saving and search +- Provides an internal chat UI. +- Manages prompt templates. +- Saves and searches conversation history. ### 2. Secure API Proxy Server (LiteLLM) -- Secure LLM access based on AWS Bedrock -- Integrated management of the Claude-3 series (Opus/Sonnet/Haiku) -- Load balancing and rate limiting of requests -- Centralized API key management +- Secure LLM access based on AWS Bedrock. +- Integrated management of the Claude-3 series (Opus/Sonnet/Haiku). +- Load balancing and rate limiting of requests. +- Centralized API key management. -### 3. Cost Management and Monitoring Infrastructure (Langfuse) -- Visualization of token usage -- Departmental cost aggregation -- Usage analysis +### 3. Cost Management and Monitoring Foundation (Langfuse) +- Visualization of token usage. +- Departmental cost aggregation. +- Usage analysis. ## 🏗️ System Architecture -### Secure 3-Tier Architecture based on AWS Bedrock +### Secure 3-Tier Architecture Based on AWS Bedrock ```mermaid %%{init:{'theme':'base'}}%% @@ -121,17 +121,18 @@ Recommended Configuration: 1. **Development Department** - Code review assistance - Bug analysis efficiency improvement - - Documentation generation + - Document generation 2. **Business Departments** - Report generation assistance - Data analysis support - - Minutes creation + - Meeting minutes creation 3. **Customer Support** - - Inquiry response efficiency improvement + - Improved inquiry response efficiency - Automatic FAQ generation - - Improvement of reply text quality + - Improved response quality + ## 🔧 Installation and Operation @@ -143,7 +144,7 @@ cd AMATERASU # 2. Set environment variables cp .env.example .env -# Edit the .env file and set credentials +# Edit the .env file and set your credentials # 3. Deploy infrastructure cd spellbook/base-infrastructure @@ -153,7 +154,7 @@ cd ../open-webui/terraform/main-infrastructure terraform init && terraform apply # 4. Start services -# Langfuse (monitoring infrastructure) +# Langfuse (monitoring foundation) cd ../../langfuse docker-compose up -d @@ -168,18 +169,26 @@ docker-compose up -d ## 📚 Detailed Documentation -- [Spellbook Infrastructure Setup Guide](spellbook/README.md) +- [Spellbook Infrastructure Construction Guide](spellbook/README.md) - [LiteLLM Configuration Guide](spellbook/litellm/README.md) - [Langfuse Setup Guide](spellbook/langfuse/README.md) ## 🆕 Latest Information +### v0.6.1 Update Notes + +- Updated documentation and added important information to the README file. +- Updated English and Japanese READMEs. +- Added information about the development process using SourceSage and claude.ai. +- Simplified security-related descriptions. + + ### v0.6.0 Update Notes - Removed unnecessary resources due to the removal of the CloudFront infrastructure. - Simplified the code to improve maintainability. - Added application HTTPS and HTTP URLs to the output. -- Made it easier to change the path of the environment variable file and setup script in `terraform.tfvars`. +- Made it easier to change the paths of the environment variable file and setup script in `terraform.tfvars`. - Removed unnecessary variable definitions. - Simplified the setup script. @@ -193,11 +202,11 @@ Provides detailed cost analysis and management features through Langfuse: ## 👏 Acknowledgements -Thanks to Maki for their contributions. +Thanks to iris-s-coon and Maki for their contributions. ## 📄 License -This project is licensed under the MIT License. See the [LICENSE](LICENSE) file for details. +This project is licensed under the MIT License. See the [LICENSE](LICENSE) file for details. ## 🤝 Contributions @@ -210,7 +219,7 @@ This project is licensed under the MIT License. See the [LICENSE](LICENSE) file ## 📧 Support For questions or feedback, please feel free to contact us: -- Create an Issue: [GitHub Issues](https://github.com/Sunwood-ai-labs/AMATERASU/issues) +- Create an issue: [GitHub Issues](https://github.com/Sunwood-ai-labs/AMATERASU/issues) - Email: support@sunwoodai.com ---