A powerful multi-agent AI system built with CrewAI that analyzes Keboola data integration and transformation projects. The crew consists of specialized agents that examine source systems, integrations, and transformations to provide comprehensive analysis and documentation.
Copy the environment template and configure your variables:
cp .env.dist .envEdit the .env file and fill in the appropriate values for all required variables. Examples are provided in the .env.dist file.
Launch your crew of AI agents:
docker-compose run crewThis command initializes the KeboolaDoc Crew using Docker, assembling the agents and assigning them tasks as defined in your configuration.
The KeboolaDoc Crew consists of two specialized AI agents:
- Data Integration Expert: Analyzes source systems and data integration patterns
- Transformation Expert: Examines data transformations and processing logic
The crew performs sequential analysis of your Keboola project and generates comprehensive reports.
All analysis results are saved to the output/ folder:
output/integrations.txt- Source system and integration analysisoutput/transformations.txt- Transformation analysis and documentation
Make sure the output/ directory exists before running the crew.
crewai-mcp/
├── src/keboola_doc/
│ ├── config/
│ │ ├── agents.yaml # Agent definitions
│ │ └── tasks.yaml # Task configurations
│ ├── crew.py # Main crew logic
│ └── main.py # Entry point
├── output/ # Analysis results
│ ├── integrations.txt # Integration analysis
│ └── transformations.txt # Transformation analysis
├── docker-compose.yml # Docker configuration
├── .env.dist # Environment template
└── README.md # This file
Feel free to contribute to this project by submitting issues or pull requests.
This project is licensed under the terms specified in the LICENSE file.
Happy AI Collaboration! 🤖✨