diff --git a/docs/README.en.md b/docs/README.en.md index ed90d25..dbbb1e7 100644 --- a/docs/README.en.md +++ b/docs/README.en.md @@ -8,17 +8,17 @@ License

-

Enterprise-Grade Private AI Platform (v1.20.0)

+

Enterprise-Grade Private AI Platform (v1.21.0)

>[!IMPORTANT] >This repository leverages [SourceSage](https://github.com/Sunwood-ai-labs/SourceSage). Approximately 90% of the release notes, README, and commit messages were 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 as an independent EC2 instance using Docker Compose, enabling easy deployment with Terraform. +>AMATERASU is the successor project to [MOA](https://github.com/Sunwood-ai-labs/MOA). It has evolved to run each AI service as an independent EC2 instance using Docker Compose, allowing for easy deployment with Terraform. ## 🚀 Project Overview -AMATERASU is an enterprise-grade private AI platform. Built on AWS Bedrock and Google Vertex AI, it allows for the development and operation of LLM-based applications in a secure and scalable environment. Integration with GitLab streamlines version control, CI/CD pipelines, and project management. This repository serves as a "spellbook" for managing multiple AI-related projects. Each project is structured as an independent folder for deploying and managing specific AI services or functionalities. +AMATERASU is an enterprise-grade private AI platform. Built on AWS Bedrock and Google Vertex AI, it enables the development and operation of LLM-based applications in a secure and scalable environment. Integration with GitLab streamlines version control, CI/CD pipelines, and project management. This repository acts as a "Spellbook" for managing multiple AI-related projects. Each project is structured as an independent folder for deploying and managing specific AI services or functionalities. ## ✨ Key Features @@ -34,19 +34,19 @@ AMATERASU is an enterprise-grade private AI platform. Built on AWS Bedrock and G - Flexible scaling ### Infrastructure as Code -- Fully automated deployment with Terraform +- Fully automated deployment using Terraform - Environment-specific configuration management -- Version-controlled configuration +- Version-controlled infrastructure ### GitLab Integration -- Enhanced version control, CI/CD pipelines, and project management capabilities +- Enhanced version control, CI/CD pipelines, and project management - Integration with self-hosted GitLab instances - LLM-powered merge request analysis - Automated labeling using GitLab webhooks -### Project Exploration Functionality +### Project Exploration Feature - Automatic detection of Terraform projects and generation of `terraform.tfvars` files -- Simplified configuration via the `amaterasu` command-line tool +- Simplified configuration using the `amaterasu` command-line tool ## 🏗️ System Architecture @@ -61,10 +61,12 @@ graph TB LF["Langfuse
Monitoring"] GL["GitLab
Version Control"] CD["Coder
Cloud Development Environment"] + GR["Gradio LLM Tester"] + ST["Streamlit LLM Tester"] end subgraph "Fargate-based Service" - PP["Prompt Pandora
Prompt Generation Assistance"] + PP["Prompt Pandora
Prompt Generation Support"] ECS["ECS Fargate Cluster"] end end @@ -85,6 +87,8 @@ graph TB LF --> CF GL --> CF CD --> CF + GR --> CF + ST --> CF PP --> ECS CF --> WAF @@ -108,6 +112,7 @@ graph TB ### 2. LiteLLM (API Proxy) - Unified access to Claude-3 series models - Access to Google Vertex AI models +- OpenRouter API integration - API key management and rate limiting - [Details here](./spellbook/litellm/README.md) @@ -120,7 +125,7 @@ graph TB ### 4. GitLab (Version Control) - Self-hosted GitLab instance - Project and code management -- CI pipeline and Runner configuration +- CI pipeline and runner configuration - Backup and restore functionality ### 5. FG-prompt-pandora (Fargate Sample Application) @@ -130,14 +135,14 @@ graph TB - [Details here](./spellbook/fg-prompt-pandora/README.md) ### 6. Coder (Cloud Development Environment) -- Web-based IDE environment +- Web-based IDE - Support for VS Code extensions - Secure development on AWS infrastructure - [Details here](./spellbook/Coder/README.md) ### 7. Dify (AI Application Development Platform) - AI application development platform integrating various AI models -- UI/API-based development possible +- UI/API-based development - [Details here](./spellbook/dify/README.md) ### 8. Dify Beta (AI Application Development Platform) @@ -146,44 +151,54 @@ graph TB - [Details here](./spellbook/dify-beta1/README.md) ### 9. Open WebUI Pipeline -- Pipeline functionality enhancing integration with Open WebUI +- Pipeline features enhancing integration with Open WebUI - Filter processing such as conversation turn limits and Langfuse integration - [Details here](./spellbook/open-webui-pipeline/README.md) ### 10. Amaterasu Tool (Terraform Variable Generator) - Automates the generation of `terraform.tfvars` files using a command-line tool - Generates configuration values for each project in the spellbook - - [Details here](./spellbook/amaterasu-tool-ui/README.md) +- [Details here](./spellbook/amaterasu-tool-ui/README.md) ### 11. Kotaemon (Document and Chat RAG UI Tool) -- RAG UI tool for interacting with documents and chat +- RAG UI tool for document and chat access - Provides Docker environment and Terraform configuration - Data persistence and customizable environment settings - Secure authentication system implemented - [Details here](./spellbook/kotaemon/README.md) ### 12. Bolt DIY (AI Chat Interface) -- State-of-the-art AI chat interface +- Modern AI chat interface - Supports multiple AI providers (OpenAI, Anthropic, Google, etc.) - Provides a Docker containerized environment -- CloudFront infrastructure setup +- CloudFront infrastructure configuration - [Details here](./spellbook/bolt-diy/README.md) +### 13. LLM Tester (Gradio version) +- Gradio-based LLM proxy connection tester +- Various parameter settings and debug information display + - [Details here](./spellbook/ee-llm-tester-gr/README.md) + +### 14. LLM Tester (Streamlit version) +- Streamlit-based LLM proxy connection tester +- Various parameter settings and debug information display + - [Details here](./spellbook/ee-llm-tester-st/README.md) + ## 🔧 Usage -Refer to the respective README files for instructions on using each component. For instructions on using the `amaterasu` command-line tool, see `spellbook/amaterasu-tool-ui/README.md`. +Refer to the respective README files for instructions on using each component. For instructions on using the `amaterasu` command-line tool, refer to `spellbook/amaterasu-tool-ui/README.md`. -## 📦 Installation Instructions +## 📦 Installation -1. Clone the repository: +1. Clone the repository. ```bash git clone https://github.com/Sunwood-ai-labs/AMATERASU.git cd AMATERASU ``` 2. Follow the instructions in each project's README to install dependencies and deploy the application. -3. Enter the necessary settings in the `terraform.tfvars` file. You can also use the `amaterasu` tool to generate them automatically. +3. Enter the necessary settings in the `terraform.tfvars` file. You can also use the `amaterasu` tool to generate it automatically. ## 📦 Dependencies @@ -208,21 +223,18 @@ Thanks to iris-s-coon and Maki for their contributions. ## 🆕 What's New -### AMATERASU v1.20.0 (Latest Release) +### AMATERASU v1.21.0 (Latest Release) - + -- New project additions: - - Kotaemon: Document and chat RAG UI tool - - Bolt DIY: State-of-the-art AI chat interface (supports OpenAI, Anthropic, Google, etc.) -- Infrastructure improvements: - - Updated `.gitignore` and `.SourceSageignore` (excluding Kotaemon application data and venv) - - Optimized Docker environment (added automatic update functionality with Watchtower) - - Enhanced LiteLLM functionality (strengthened Langfuse integration) -- Updated configuration files: Improved Docker Compose, environment variable templates, and Terraform modules -- Updated documentation: Improved README files (including English translation), installation instructions, and setup guides +- **Expanded LLM Tester Functionality**: Added two interfaces based on Gradio and Streamlit. Displays IP address information and detailed debugging information. +- **Enhanced LiteLLM Functionality**: Integrated OpenRouter API and standardized model names. +- **Infrastructure Improvements**: Optimized Docker environment, strengthened inter-container communication, and updated NextAuth URL settings for Langfuse3. +- **Addition of New Presentation Tool**: Added Marp Editable UI. Provides a containerized Markdown presentation editing environment. +- **Enhanced Security**: Improved WAF settings, CSV file management for IP whitelists, and flexible security rule settings. +- **Improved Configuration Management**: Added `.env.example` files for each component to standardize and document configuration files. -- Notes: - - Be sure to change the initial authentication information for Kotaemon. - - Properly manage API keys and avoid committing them to public repositories. - - Refer to the respective READMEs for newly added projects. \ No newline at end of file +- 📝 Notes: + - Existing configuration files need to be updated due to LiteLLM configuration changes. + - The WAF IP whitelist needs to be migrated to the new CSV format. + - Check environment variable settings and update as needed. \ No newline at end of file