ADK Core

Initialize Google ADK projects and agents with Python or YAML configurations.

3|Updated Jan 9, 2026
One-click install
npx skills add https://github.com/MattMagg/agentic-plugins --skill adk-core
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ADK Core
Source: https://github.com/MattMagg/agentic-plugins/tree/main/plugins/adk-builder/skills/adk-core
Command: npx skills add https://github.com/MattMagg/agentic-plugins --skill adk-core

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill guides users through initializing Google ADK projects and creating agents, reducing setup time and ensuring proper structure and authentication.

Core Features & Use Cases

  • Project scaffolding with adk create to generate Python or YAML configurations for new projects.
  • Authentication setup guidance for Google Studio and Vertex AI, including environment configuration.
  • Guidance on choosing between Python code and YAML declarative configurations, plus examples of creating LlmAgent or custom agents.

Quick Start

Run adk create to scaffold a new ADK project and then run adk run to start your first agent.

Frequently Asked Questions about ADK Core

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I initialize a new Google ADK project?

You can scaffold a new ADK project using the `adk create` command, which generates the necessary project structure and configuration files to start building LLM agents.

What's the best way to configure authentication for Google ADK agents?

Configuring authentication for Google ADK involves setting up environment configurations for Google Studio or Vertex AI to ensure your LLM agents can securely access model endpoints.

Should I use Python code or YAML declarative configuration for ADK agents?

Choosing between Python code and YAML declarative configuration depends on your workflow preference; ADK supports scaffolding both custom LlmAgent implementations and declarative YAML setups.

How do I start an LLM agent after project setup is complete?

After completing the project setup and initialization, you can start your first LLM agent by running the `adk run` command within the generated ADK project directory.

Can I use Google ADK to scaffold both custom agents and LlmAgent configurations?

Yes, Google ADK scaffolding supports creating both standard LlmAgent setups and custom agents, providing guidance and references for initializing either Python or YAML configurations.