CUGA is an open-source generalist agent for the enterprise, supporting complex task execution on web and APIs, OpenAPI/MCP integrations, composable architecture, reasoning modes, and policy-aware features. It is under active development, with minor versions released frequently introducing new features and improvements.
pip install cugaVerified import paths — ran on the pinned version, not inferred.
This quickstart demonstrates how to initialize a `CugaAgent` and invoke it with a simple task. It relies on environment variables (like `OPENAI_API_KEY`) for LLM configuration, which can be loaded from a `.env` file using `python-dotenv`.
Always review release notes for new versions and pin your `cuga` dependency to a specific minor version in production environments (e.g., `cuga==0.2.*`).
Ensure your Python environment is within the supported range (>=3.10, <3.14). Use `pyenv` or `conda` to manage Python versions if necessary.
Set the required LLM API keys in your environment variables. It is recommended to use a `.env` file and `python-dotenv` for local development, or your cloud provider's secrets management for deployment.