ADK Recipes is a collection of sample agents built with the Agent Development Kit that demonstrates how to solve real problems using ADK.
The project addresses the challenge of starting agent development from scratch by providing working, runnable examples that teach specific patterns and approaches. Rather than beginning with a blank page, developers can fork a recipe as a foundation for their own project or study the patterns to understand how to implement features like OAuth flows, session memory, guardrails, and RAG patterns. The recipes are organized into two categories: core recipes maintained by the agents-cli team that focus on canonical patterns and teach one concept well, and community-contributed recipes that cover broader use cases and industry-specific workflows.
Developers building customer service bots, research agents, or other AI-powered applications should consider this collection if they want a working starting point rather than building from fundamentals. Each recipe includes its own setup and run instructions. The project supports multiple language SDKs including Python, TypeScript, Go, Java, and Kotlin, so teams can choose their preferred language. The collection is explicitly positioned for demonstration and learning purposes rather than production deployment.
The project maintains a structured approach to contributions through documented guidelines including a Recipe Checklist and Recipe Handbook, indicating organized processes for accepting new recipes. The repository includes tooling for recipe scaffolding and manifest generation within its agents skills directory, suggesting active investment in making it easier for contributors to add new examples. Community contributions are actively solicited and integrated alongside the curated core recipes.
