This project provides a framework for building AI-powered social media bots, specifically targeting Bilibili. It enables users to automate interactions like live stream chat responses, video comment replies, and direct messaging, leveraging LLMs for conversational capabilities. The target audience includes content creators and developers looking to automate social media engagement.

How It Works

Langup utilizes a modular architecture, with distinct "Uploader" classes for different bot functionalities (e.g., VtuBer for live streams, VideoCommentUP for video comments, ChatUP for DMs). These modules integrate with LLM providers (like OpenAI) and Bilibili's API to process incoming messages, generate responses based on user-defined system prompts, and post them back to the platform. The design emphasizes flexibility, allowing customization of AI personas, response strategies, and interaction parameters.

Quick Start & Requirements

• Install via pip: pip install langup==0.0.10
• Or clone and install dependencies: git clone https://github.com/jiran214/langup-ai.git && cd langup-ai/ && python -m pip install -r requirements.txt
• Requires Python >= 3.8.
• OpenAI API key is necessary.
• Bilibili authentication credentials (sessdata, buvid3, bili_jct) can be provided directly, via .env file, or automatically read from browser sessions (Edge recommended).
• Official documentation: https://nemo2011.github.io/bilibili-api/#/get-credential

Highlighted Details

• Supports Bilibili live stream chat bots with customizable personas (e.g., "contrarian" replies).
• Automates replies to video comments that mention specific keywords.
• Includes a direct messaging bot and a real-time voice interaction assistant.
• Offers configuration for LLM parameters, proxy settings, and logging.

Maintenance & Community

The project lists Langchain and Bilibili API as dependencies. Further community engagement channels are not explicitly mentioned in the README.

Licensing & Compatibility

The project's licensing is not explicitly stated in the README. Users are cautioned against misuse and advised to comply with platform-specific terms of service.

Limitations & Caveats

The README notes that some modules are still under development. Automatic browser credential reading may require closing all browser processes. The effectiveness of prompts and filtering is dependent on Bilibili's platform policies.