Pike is an agentic platform for building and interacting with personalized AI agents. Users can create specialized agents tailored to unique knowledge domains and engage in dynamic conversations through a responsive web interface powered by a Python backend.
- Create and manage specialized AI agents
- Chat interface for personalized exploration
- Modular architecture with Python backend and modern JavaScript frontend
- Easy-to-use development and execution scripts
pike/
│
├── frontend/ # Frontend application (npm-based)
├── backend/ # Backend application (Python-based, entry: pike.py)
├── setup.sh # Script to install dependencies and prepare environment
├── start.sh # Script to launch frontend and backend
├── .env # Environment file for API keys (you must create this)
└── README.md
-
Clone the repository and enter the project directory.
-
Create a
.envfile in the root directory with the following content:GOOGLE_API_KEY=your_api_key_here
You can obtain your API key by following the instructions here:
https://ai.google.dev/gemini-api/docs/api-key -
Run the setup script:
bash setup.sh
This will:
- Install Node.js using NVM
- Install frontend dependencies
- Create a Python virtual environment
- Install Python backend dependencies
Once setup is complete, launch the platform using:
bash start.shThis script will:
Start the backend server from backend/pike.py
Start the frontend development server from frontend/
Activate the Python virtual environment automatically
You can then access the platform in your browser at:
- Unix-like shell environment (Linux/macOS)
- Python 3.8+
- Bash
- Git
- curl
The setup script will install Node.js via NVM if it's not already installed.
- To manually activate the backend environment:
source backend/.venv/bin/activate
This project is licensed under the MIT License.
