Pippin-Lite is a lighter-weight cousin of the Pippin framework. At 227 lines of code, it’s light-weight, approachable, and designed as an educational starting point for exploring dynamic self-building AI agents. While the core Pippin framework focuses on complex and expansive capabilities, Pippin-Lite keeps things simple and close-ended, making it perfect for exploring tasks with close-ended goals.
Unlike its robust sibling, Pippin-Lite is intentionally minimalistic. It’s focused on helping users understand the fundamentals of autonomous agents and how they build and register tools dynamically. With potential enhancements like memory and self-improvement, it offers a flexible foundation to expand your AI projects.
We’d like to thank the amazing Composio team for their help integrating Composio tools into this project!
Join our Pippin Developer Hub on Discord to connect with other developers, share ideas, and collaborate on exciting Pippin-inspired projects.
- Execution Safety: Pippin-Lite installs dependencies and executes code dynamically. Run it in a safe environment and be mindful of the tasks you define.
- Privacy: API keys and sensitive data should be securely managed.
- Dynamic Tool Creation: Automatically creates and registers tools to accomplish user-defined tasks.
- Composio Integration: Enables advanced workflows and seamless app integration using Composio tools.
- Authentication Management: Detects available API keys dynamically for various integrations.
- Package Installation: Automatically installs required Python packages during tool creation.
- Error Handling: Gracefully handles errors and iterates to complete tasks.
- Dynamic Execution: Iteratively builds, registers, and utilizes tools during runtime.
- Minimalistic Design: Designed as a lightweight introduction to autonomous agent workflows.
- Python 3.8 or higher
pippackage manager
-
Clone the Repository
git clone https://github.com/pippinlovesyou/pippin-lite.git cd pippin-lite -
Create a Virtual Environment (Optional)
python3 -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate
-
Install Dependencies
pip install -r requirements.txt
-
Set Environment Variables
- Create a
.envfile in the project directory and add your API keys:LITELLM_MODEL=gpt-4o OPENAI_API_KEY=your-openai-api-key
- Create a
-
Run the Application
python pippin-lite.py
-
Describe the Task When prompted, enter a task description. The agent will dynamically create tools and attempt to solve the task.
-
Monitor Progress Progress updates will be displayed in the console. Once the task is completed, you will see a "Task completed" message.
- Summarize news headlines from a website.
- Analyze an image in your folder (requires an image file).
- Search Twitter for a keyword and summarize findings.
- Dynamic Toolset: Uses Composio tools for workflows and dynamic action execution.
- LiteLLM Integration: Manages AI interactions via LiteLLM.
- Error Recovery: Automatically retries failed tool executions.
- Self-Building Agent: Dynamically creates tools to handle unforeseen tasks.
We encourage contributions to improve Pippin-Lite and its integration into the Pippin framework. Fork the repo, make changes, and submit a pull request. Contributions to Pippin-Universe are also welcome to share your own Pippin-inspired creations.
- Enhance tool creation workflows.
- Add support for new APIs or models.
- Extend functionality for more complex tasks.
Pippin-Lite is licensed under the MIT License. See the LICENSE file for details.
Happy building with Pippin! 🦄✨