dotsagent.io
Language:English
Framework · Google

Google ADK at a glance

Google's Agent Development Kit, with SDKs for Python, TypeScript, Go, Java and Kotlin and evaluation built into the core. It is Gemini-first but reaches other models through adapters, and deploys to Agent Runtime on Google Cloud.

Best for: Teams working across several languages that want Gemini-first agents, built-in evals and a direct route to Google Cloud.

Key facts

Maintainer
Google
Languages
Python, TypeScript, Go, Java, Kotlin
Licence
Apache-2.0
Latest version
Python 2.10.0 · September 25, 2026
Repository
google/adk-python · 21,687 stars
Hosted option
Agent Runtime on Gemini Enterprise Agent Platform (formerly Vertex AI Agent Engine); also Cloud Run, GKE
Agent Runtime was formerly Vertex AI Agent Engine; the deploy call is still client.agent_engines.create(...).

Quickstart

Install the package, set the credentials, then run the smallest agent from the official docs. Model IDs are the ones the docs use and may not be enabled on your account.

Python

Install
pip install google-adk && adk create my_agent  (run with: adk run my_agent  or  adk web --port 8000)

Needs GOOGLE_API_KEY in my_agent/.env

Python
from google.adk.agents.llm_agent import Agent

# Mock tool implementation
def get_current_time(city: str) -> dict:
    """Returns the current time in a specified city."""
    return {"status": "success", "city": city, "time": "10:30 AM"}

root_agent = Agent(
    model='gemini-flash-latest',
    name='root_agent',
    description="Tells the current time in a specified city.",
    instruction="You are a helpful assistant that tells the current time in cities. Use the 'get_current_time' tool for this purpose.",
    tools=[get_current_time],
)

Quickstart source

TypeScript

Install
npm install @google/adk (inferred from package name; not copied from docs)

Needs GOOGLE_API_KEY

TypeScript
import { LlmAgent, GOOGLE_SEARCH } from '@google/adk';

const agent = new LlmAgent({
  name: 'researcher',
  model: 'gemini-flash-latest',
  instruction: 'You help users research topics thoroughly.',
  tools: [GOOGLE_SEARCH],
});

Quickstart source

Features

Any modelYesGemini first; other models through LiteLLM and local adapters
MCP clientYesMcpToolset, with optional confirmation per tool
MCP serverYesExpose an agent with to_mcp_server()
Multi-agentYesSub-agent trees, plus graph workflows since ADK 2.0
Durable runsYesResume stopped agents; Agent Runtime keeps long-running state
Human approvalYesTool Confirmation via require_confirmation=True
TracingYesOpenTelemetry-based, with Cloud Trace on Google Cloud
EvalsYesEvaluation is a core ADK feature

Recent changes

  • ADK 2.0 added graph-based workflows.
  • A Kotlin SDK joined Python, TypeScript, Go and Java.
  • The docs moved to adk.dev.
  • Vertex AI Agent Engine was renamed Agent Runtime on Gemini Enterprise Agent Platform.

Alternatives in the same language

Sources

  1. google.github.io/adk-docs/
  2. google.github.io/adk-docs/get-started/python/
  3. adk.dev/tools-custom/confirmation/
  4. adk.dev/tools-custom/mcp-tools/advanced/
  5. adk.dev/tools-custom/mcp-tools/agent-as-server/
  6. docs.cloud.google.com/gemini-enterprise-agent-platform/vertex-ai-name-changes
  7. cloud.google.com/blog/products/ai-machine-learning/introducing-gemini-enterprise-agent
  8. docs.cloud.google.com/gemini-enterprise-agent-platform/build/runtime/quickstart-adk
  9. pypi.org/pypi/google-adk/json
  10. registry.npmjs.org/@google%2fadk
  11. api.github.com/repos/google/adk-python

Independent reference for people who build AI agents. Not affiliated with any vendor named here.

© 2026 DotsAgent · Facts checked October 1, 2026