This project provides a framework for building and deploying AI-powered chatbots across various platforms, including Telegram, Slack, Gmail, and websites. It targets developers and users who want to integrate large language models like ChatGPT, Claude, and Gemini into their applications with minimal configuration, offering a unified approach to model and channel management.

How It Works

The core architecture uses a flexible configuration-driven approach. Users define a config.json file to specify the AI model (e.g., OpenAI's GPT) and the target application channel (e.g., Slack, Telegram). The system dynamically loads and connects these components, allowing for easy switching between different models and channels within a single project. This design promotes reusability, enabling new models to be deployed across all supported channels and existing channels to leverage new model capabilities.

Quick Start & Requirements

• Install: git clone https://github.com/zhayujie/bot-on-anything followed by pip3 install -r requirements.txt.
• Prerequisites: Python 3.7-3.10, Linux, macOS, or Windows. Specific channels may require additional dependencies (e.g., openai, slack_bolt, werobot, aiocqhttp).
• Configuration: Copy config-template.json to config.json and update model and channel-specific API keys and settings.
• Run: python3 app.py (default channel is terminal).
• Docs: English, 中文

Highlighted Details

• Supports multiple AI models including ChatGPT and LinkAI.
• Integrates with numerous communication channels: Terminal, Web, WeChat (MP & Service), Telegram, QQ, DingTalk, Feishu, Gmail, Slack.
• Allows customization of AI model behavior via parameters like character_desc, temperature, and top_p.
• WeChat MP channel has an optimized 15-second timeout for responses, mitigating the 5-second limit.

Maintenance & Community

The project lists several contributors. Community interaction channels are not explicitly mentioned in the README.

Licensing & Compatibility

The README does not specify a license. Compatibility for commercial use or closed-source linking is not detailed.

Limitations & Caveats

The QQ channel requires running a separate go-cqhttp process. WeChat MP has a 15-second response limit, which may still be insufficient for complex AI queries. The project does not specify a license, which could impact commercial adoption.