This project provides a question-answering system built on a vector knowledge base derived from the "Xianzhi Community" (先知社区) articles, aimed at cybersecurity professionals and researchers. It enables quick retrieval and summarization of relevant information from a large corpus of security-focused content, aiding in tasks like CTF competitions.

How It Works

The system constructs a vector knowledge base from specified ranges of Xianzhi Community articles. It leverages Large Language Models (LLMs) to process user queries, find the most relevant documents within the vector store, and generate synthesized answers. This approach allows for efficient semantic search and contextual understanding of cybersecurity topics.

Quick Start & Requirements

• Install dependencies: pip install -r requirements.txt
• Configure API keys and model providers in .env (e.g., GOOGLE_API_KEY, OPENAI_API_KEY).
• Requires Python 3.8+.
• Supports Gemini, OpenAI, and Ollama model providers.
• Local vector store directory is configurable.

Highlighted Details

• Supports local document ingestion and URL-based querying.
• Allows updating the knowledge base with local documents.
• Offers flexible model configuration via .env for Gemini, OpenAI, and Ollama.
• Prioritizes command-line arguments over environment variables for model selection.

Maintenance & Community

• The project is maintained by kento996.
• No specific community links (Discord/Slack) or roadmap are provided in the README.

Licensing & Compatibility

• The README does not explicitly state a license.
• Usage is restricted to research purposes; commercial or illegal use is prohibited.

Limitations & Caveats

The project is intended for research use only and explicitly warns against illegal activities. The scope of the knowledge base is limited to articles 7023-12923 from the Xianzhi Community.