diff --git a/docs/README.en.md b/docs/README.en.md index da3fcf8..459dd9e 100644 --- a/docs/README.en.md +++ b/docs/README.en.md @@ -3,48 +3,38 @@

🌟 swarm-sample-box 🌟

-

- - GitHub Repo - - - License - - - GitHub release - - - PRs Welcome - -

- - +[![GitHub Repo](https://img.shields.io/badge/github-swarm__sample__box-blue?logo=github)](https://github.com/Sunwood-ai-labs/swarm-sample-box) +[![License](https://img.shields.io/github/license/Sunwood-ai-labs/swarm-sample-box?color=green)](https://github.com/Sunwood-ai-labs/swarm-sample-box/blob/main/LICENSE) +[![GitHub stars](https://img.shields.io/github/stars/Sunwood-ai-labs/swarm-sample-box?style=social)](https://github.com/Sunwood-ai-labs/swarm-sample-box/stargazers) +[![GitHub release](https://img.shields.io/github/v/release/Sunwood-ai-labs/swarm-sample-box?include_prereleases&style=flat-square)](https://github.com/Sunwood-ai-labs/swarm-sample-box/releases) +... (Other badges are similarly linked and self-explanatory)

～ Experimental Playground for AI Agent Orchestration ～ - JA doc - EN doc + [![JA doc](https://img.shields.io/badge/ドキγƒ₯γƒ‘γƒ³γƒˆ-ζ—₯本θͺž-white.svg)](https://github.com/Sunwood-ai-labs/swarm-sample-box/blob/main/README.md) + [![EN doc](https://img.shields.io/badge/english-document-white.svg)](https://github.com/Sunwood-ai-labs/swarm-sample-box/blob/main/docs/README.en.md)

> [!IMPORTANT] -> swarm-sample-box is a template repository developed using [cline (formerly Claude Dev)](https://github.com/clinebot/cline), [SourceSage](https://github.com/Sunwood-ai-labs/SourceSage), and [claude.ai](https://claude.ai/). The majority of the release notes, README, and commit messages have been generated using state-of-the-art AI technology. +> swarm-sample-box is a template repository developed using [cline (formerly Claude Dev)](https://github.com/clinebot/cline), [SourceSage](https://github.com/Sunwood-ai-labs/SourceSage), and [claude.ai](https://claude.ai/). Most of the release notes, README, and commit messages are generated using the latest AI technologies. ## πŸš€ Project Overview -Swarm Sample Box is an AI agent experimentation repository utilizing [Swarm, an experimental multi-agent orchestration framework developed by OpenAI](https://github.com/openai/swarm). This repository provides a lightweight and flexible solution for efficiently coordinating multiple AI agents to perform complex tasks, supporting developers' research and experimentation. Version: v1.5.0 +Swarm Sample Box is an AI agent experimental repository that utilizes [Swarm, an experimental multi-agent orchestration framework developed by OpenAI](https://github.com/openai/swarm). This repository provides a lightweight and flexible solution for efficiently coordinating multiple AI agents to perform complex tasks, supporting developers' research and experimentation. Version: v1.3.1 ## ✨ Main Features -- Diverse Agent Samples: Provides implementation examples of various AI agents, from basic conversations to complex task processing. -- Flexible Customization: Each sample can be easily extended and modified to adapt to your own use cases. -- Integrated Experimental Environment: Offers a practical experimental environment, including Docker environment and integration with a vector database (Qdrant). -- Automatic Evaluation Function: Some samples include automatic evaluation scripts to measure agent performance. -- Gemini Pro 002 model support +1. **Diverse Agent Samples**: Provides implementation examples of various AI agents, ranging from basic dialogue to complex task processing. +2. **Flexible Customization**: Each sample can be easily extended and modified, adaptable to your own use cases. +3. **Integrated Experimental Environment**: Provides a practical experimental environment, including integration with Docker environments and vector databases (Qdrant). +4. **Automatic Evaluation Function**: Some samples include automatic evaluation scripts, allowing for agent performance measurement. +5. πŸŽ‰ **Sample Question Display Function**: Added a function to display sample questions to the user. + -## πŸ”§ Usage +## πŸ”§ Setup and Usage 1. Clone the repository: ```bash @@ -64,66 +54,66 @@ Swarm Sample Box is an AI agent experimentation repository utilizing [Swarm, an pip install -r requirements.txt ``` -4. Set your OpenAI API key: +4. Set the OpenAI API key: ```bash export OPENAI_API_KEY="your-api-key" ``` 5. Run a specific sample: ```bash - cd examples/ + cd examples/ python main.py ``` +## πŸ“š Sample List -## πŸ“¦ Installation Instructions +### Official Samples (Translated to Japanese) -The `swarm` package needs to be installed from the GitHub repository, not PyPI, as specified in `requirements.txt`. Therefore, please install it using pip as follows: +The following samples are Japanese translations of the samples listed in the [official OpenAI Swarm repository](https://github.com/openai/swarm/tree/main/examples), modified to be operational: -```bash -pip install -r requirements.txt -# or -pip install git+https://github.com/openai/swarm.git -``` +- [Basic](https://github.com/Sunwood-ai-labs/swarm-sample-box/tree/main/examples/basic): A minimal implementation example introducing the basic functions of Swarm. +- [Airline](https://github.com/Sunwood-ai-labs/swarm-sample-box/tree/main/examples/airline): A multi-agent setup simulating an airline's customer service. +- [Personal Shopper](https://github.com/Sunwood-ai-labs/swarm-sample-box/tree/main/examples/personal_shopper): An implementation example of a personal shopping assistant agent. +- [Support Bot](https://github.com/Sunwood-ai-labs/swarm-sample-box/tree/main/examples/support_bot): An implementation example of a customer support bot, including a document search function using Qdrant. +- [Triage Agent](https://github.com/Sunwood-ai-labs/swarm-sample-box/tree/main/examples/triage_agent): An implementation example of a triage agent that directs user requests to the appropriate agent. +- [Weather Agent](https://github.com/Sunwood-ai-labs/swarm-sample-box/tree/main/examples/weather_agent): A simple implementation example of an agent that retrieves weather information and performs related tasks. (πŸ”₯Under Development) +- [Customer Service Streaming](https://github.com/Sunwood-ai-labs/swarm-sample-box/tree/main/examples/customer_service_streaming): A sample simulating customer service in a streaming format. (πŸ”₯Under Development) -## πŸ†• What's New - -- The README for the Triage Agent sample has been significantly improved, adding a workflow diagram, setup instructions, evaluation tests, file structure, contribution guidelines, and license information. -- Added Gemini Pro 002 model support to the Triage Agent sample. +### Original Samples +This section will introduce originally developed samples. Currently under preparation, so please wait for future updates. -## πŸ“š Sample List +- [In preparation] New original sample 1: Details coming soon +- [In preparation] New original sample 2: Details coming soon -### Official Samples (Translated to Japanese) +## πŸ§ͺ Evaluation Method -- Basic: A minimal implementation example showcasing basic Swarm functionality. -- Airline: A multi-agent setup simulating an airline's customer service. -- Personal Shopper: An implementation example of a personal shopping assistant agent. -- Support Bot: An implementation example of a customer support bot, including a document search function using Qdrant. -- Triage Agent: An implementation example of a triage agent that directs user requests to the appropriate agent. -- Weather Agent: A simple agent implementation example that retrieves weather information and performs related tasks (under development). -- Customer Service Streaming: A sample simulating customer service in a streaming format (under development). +Some samples include automatic evaluation scripts. To run the evaluation, execute the following command in the sample directory: -### Original Samples +```bash +pytest evals.py +``` -This section will introduce original samples that we have developed. Currently under preparation, so please wait for future updates. +## πŸ“¦ Upgrade Procedure -- [In preparation] New Original Sample 1: Details coming soon. -- [In preparation] New Original Sample 2: Details coming soon. +1. Clone or pull the latest version of the repository. +2. Update dependencies with `make install`. +3. Restart the Docker container with `docker-compose up -d`. (For Support Bot, also run `make prep`) +4. Run the data preparation script with `make prep` in the relevant sample. +5. Run the application with `make run`. ## 🀝 Contribution -Swarm Sample Box welcomes contributions from the community as an open-source project. We welcome any kind of contribution, such as adding new samples, improving existing samples, and expanding documentation. - +Swarm Sample Box welcomes contributions from the community as an open-source project. We welcome contributions in any form, such as adding new samples, improving existing samples, and expanding documentation. ## πŸ“„ License -Swarm Sample Box is released under the [MIT License](LICENSE). +Swarm Sample Box is published under the [MIT License](LICENSE). ## πŸ™ Acknowledgements -This project is based on the Swarm framework developed by OpenAI. We express our sincere gratitude to the Swarm developers. Contributors to this release: Maki, iris-s-coon +This project is based on the Swarm framework developed by OpenAI. We express our sincere gratitude to the developers of Swarm. Contributors to this release: Maki, iris-s-coon ---