Description
BeeBot is an Autonomous AI Agent hosted on GitHub, designed to autonomously perform a wide range of practical tasks. The project is currently on hold, as the developer believes current LLMs are not yet capable of generalized autonomous AI. The project may be revived if LLMs improve significantly, a custom model is developed, or a specific subset of tasks is identified.
BeeBot's core functionality includes tool selection via AutoPack, the ability to acquire more tools during task execution, built-in persistence, and a REST API conforming to the e2b standard. It also features a websocket server for event publishing, swappable filesystem emulation, and a web UI (coming soon). The project emphasizes functionality, flexibility, and reliability, prioritizing these aspects over developer convenience. The development process is guided by a specific philosophy, focusing on tool selection, LLM specialization, and unorthodox methodologies to increase development speed.
BeeBot's architecture and future plans are detailed in the docs/ directory. The project uses AutoPack as the package manager for its tools and leverages various LLMs. The project's priorities are functionality, flexibility, reliability, efficiency, and convenience. The project uses end-to-end tests instead of unit tests. The project is open-source and welcomes contributions and feedback from the community. The project is written primarily in Python.
To get started, users can clone the repository and install dependencies using poetry. Persistence is required, with Postgres via Docker recommended. The CLI can be run using `poetry run beebot API`. The server can be started using `uvicorn`. The API can be called using curl commands. A websocket connection is available at the /notifications endpoint. The project is actively seeking contributions and feedback.
GitHub's Core Features
Autonomous task execution
Tool selection via AutoPack
Built-in persistence
REST API conforming to e2b standard
Websocket server for event publishing
Swappable filesystem emulation
Web UI (coming soon)
Dynamic history manipulation
Built-in caching with Helicone
Focus on functionality and flexibility
Uses end-to-end tests
Open-source project
Getting Started with GitHub
Developer: Clone the repository using git clone https://github.com/AutoPackAI/beebot.git
Developer: Navigate to the project directory using cd beebot
Developer: Install dependencies using poetry install
Developer: Configure persistence (Postgres via Docker recommended)
Developer: Run the CLI using poetry run beebot API
Developer: Start the server using uvicorn beebot.initiator.api:create_app --factory --timeout-keep-alive=300
Developer: Interact with the API using curl commands to create and execute tasks
GitHub's Use Cases
- Automated Task Execution
- AI Assistant Development
- Tool Integration
- API Interaction
- Websocket Notifications
- Experimentation with AI
- Workflow Automation
- Research and Development








