This project is a Python-based AI agent that uses Google’s Gemini model to solve user prompts by reasoning step-by-step and calling external functions when needed.
Instead of only generating text, the agent can:
- Interpret a user’s request
- Decide when to call tools/functions
- Execute those functions in Python
- Feed the results back into the model
- Continue reasoning until a final answer is produced
- Uses an LLM to assist with coding tasks
- Can call tools/functions
- Handles prompts and responses
- Clone the repo
- Install dependencies
- Add your API key to environment variables
- Run the project
uv run main.py "your prompt" --verbose # for additional info, optional.- Be careful giving the agent filesystem or interpreter access
- Remove any paid API keys when finished