Agent frameworks
google/adk-python
Google’s Agent Development Kit is a code-first framework for composing, evaluating and deploying agents. It supports multiple agent patterns and connects naturally with Google’s model and cloud ecosystem while retaining an open-source development surface.
THE PRACTICAL EXPLANATION
Google’s Agent Development Kit is a code-first framework for composing, evaluating and deploying agents. It supports multiple agent patterns and connects naturally with Google’s model and cloud ecosystem while retaining an open-source development surface.
WHERE TO USE IT
It fits organisations building around Gemini, Google Cloud or existing Google services, as well as teams comparing provider-backed agent SDKs. It is useful where deployment and evaluation need to be considered alongside local development.
WHO MAY USE IT
Python developers, Google Cloud teams and enterprise architects assessing a governed route from agent prototype to hosted service.
HOW TO USE IT
Begin with a local agent and a small evaluation set. Test the exact model and deployment target early because local success does not prove cloud behaviour, latency or cost. Make provider-specific services an explicit choice and isolate them behind clear interfaces where portability matters.
GETTING THE BEST RESULTS
WHY IT MAY BE USEFUL
Code-first toolkit for building, evaluating and deploying sophisticated agents.
Best considered for: Teams building agents around Google models and services.
READ BEFORE YOU ADOPT IT
Cloud and model choices can shape portability; test the deployment path early.
Confirm the current licence, maintenance status, dependency risk, data path, model access, tool permissions and human approval points at the source. A public repository is inspectable raw material—not proof that a system is secure, supported or suitable for your production environment.