OpenClaw Assistant is a comprehensive AI-powered assistant project. It integrates various functionalities including web search, research paper retrieval (via arXiv), document parsing (PDFs), and a Retrieval-Augmented Generation (RAG) system. The project also features a Telegram bot interface, an API powered by FastAPI, and a task scheduler.

• API: Built with FastAPI and Uvicorn for serving requests.
• Telegram Bot: Integration with the python-telegram-bot library to interact with users via Telegram.
• Tools:
  • Web Search: Uses requests and beautifulsoup4 for web scraping (e.g., DuckDuckGo searches).
  • arXiv Downloader: Leverages feedparser to query and download research papers from arXiv.
  • PDF Parser: Uses pypdf to extract text from documents.
• RAG (Retrieval-Augmented Generation):
  • Vector embeddings via sentence-transformers.
  • Vector database integration using chromadb.
• Scheduler: Background task scheduling and reminders.
• Agents/Core logic: Modular architecture for AI agents and core application logic.

• /api - FastAPI endpoints and server logic.
• /agents - AI agents and prompt handling.
• /core - Core business logic and configurations.
• /data - Directory for storing local files like PDFs or databases.
• /rag - Retrieval-Augmented Generation pipelines and database interactions.
• /scheduler - Cron jobs, background tasks, and reminder services.
• /telegram - Handlers and setup for the Telegram bot interface.
• /tools - Utility scripts (e.g., web_search.py, arxiv_downloader.py, pdf_parser.py).

1. Clone the repository.
2. Create a virtual environment:

   python -m venv .venv
   source .venv/bin/activate  # On Windows use: .venv\Scripts\activate

3. Install the required dependencies:

   pip install -r requirements.txt

The project uses python-dotenv for environment variable management. Create a .env file in the root directory and add any necessary configuration values (such as API keys, Telegram Bot Tokens, etc.).

(Provide specific commands based on your main entry points, e.g.:)

To run the API:

uvicorn api.main:app --reload

To run the Telegram Bot:

python telegram/bot.py