A large collection of open-source AI agents, agent skills, and RAG apps built on Claude, Gemini, GPT, DeepSeek, Llama, and Qwen models.
google/adk-recipes
A collection of runnable agent recipes, reference patterns, and vertical plugins built with Google's Agent Development Kit (ADK).
- Stars
- 10,497
- Forks
- 2,927
- Open issues
- 41
- Last commit
- 20 hours ago
What is it?
ADK Recipes is a set of small, runnable agents that show how to solve concrete problems with the Agent Development Kit. Each recipe is meant to be forked as a starting point or read to learn a specific pattern.
The repository is organized into three areas. core/ holds focused recipes that each teach one pattern, such as OAuth flows, session memory, guardrails, and RAG. contrib/ contains community-contributed, deployable vertical agents, for example retail product search and virtual try-on, which can run on Agent Engine, Cloud Run, or Agent Garden. plugins/ contains domain-specific skill plugins packaged as SKILL.md and EVAL.yaml files for AI coding assistants.
Each recipe includes its own README with setup and run instructions. The project covers ADK Python, TypeScript, Go, Java, and Kotlin SDKs. It is provided for demonstration and as a starting point, not for production use.
Key features
- 01Focused core recipes covering OAuth, session memory, guardrails, and RAG patterns.
- 02Deployable contrib agents for retail product search and virtual try-on.
- 03Installable skill plugins with SKILL.md and EVAL.yaml files for AI coding assistants.
- 04Per-recipe README files with setup and run instructions.
- 05Examples across ADK Python, TypeScript, Go, Java, and Kotlin SDKs.
Use cases
- Learning a specific ADK pattern such as session memory or guardrails by reading a small example.
- Forking a recipe as the starting point for a customer service or research agent.
- Deploying a retail search or virtual try-on agent to Cloud Run or Agent Engine.
- Installing a domain skill plugin into an AI coding assistant.
Quick start
- 1
Install ADK using the ADK Get Started guide
- 2
Browse recipes in core/ and open the recipe's own README for setup and run instructions
Things to know
- Recipes are for demonstration and as starting points, not production use.
- Not an officially supported Google product.
- Each recipe has its own setup, so there is no single install command for the whole collection.
Topics
FAQ
- What is adk-recipes?
- A collection of runnable agent recipes, reference patterns, and vertical plugins built with Google's Agent Development Kit (ADK). ADK Recipes is a set of small, runnable agents that show how to solve concrete problems with the Agent Development Kit. Each recipe is meant to be forked as a starting point or read to learn a specific pattern.
- Is adk-recipes free to use? What license does it have?
- adk-recipes is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
- Is adk-recipes still maintained?
- Actively maintained. The last commit to adk-recipes was on Oct 19, 2026.
- What can I use adk-recipes for?
- Learning a specific ADK pattern such as session memory or guardrails by reading a small example. Forking a recipe as the starting point for a customer service or research agent. Deploying a retail search or virtual try-on agent to Cloud Run or Agent Engine. Installing a domain skill plugin into an AI coding assistant.
- How do I install adk-recipes?
- Install ADK using the ADK Get Started guide Browse recipes in core/ and open the recipe's own README for setup and run instructions
- Where can I get adk-recipes?
- The source code and installation instructions are on GitHub: https://github.com/google/adk-recipes
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source Python framework for building production-ready LLM applications with modular pipelines and agent workflows for RAG, search and agents.
A collection of 50+ runnable Jupyter notebook tutorials covering GenAI agent techniques, from simple chatbots to multi-agent systems.
A visual platform for building, testing and deploying AI agents and workflows, with built-in API and MCP server exposure.
A memory layer for AI agents and apps that stores and retrieves user, session, and agent context across conversations.
A Python framework for building RAG, agent and document-processing applications that connect LLMs to your own data.