ByteZone
Back to listClaim this project

googleDeveloper tools

Adk Python

adk-python

Google’s open-source Python framework for defining, orchestrating, evaluating, and deploying single- and multi-agent workflows.

  • Backend
  • Full stack
  • AI coding
  • Testing
  • CLI
  • API/SDK
  • Library/framework
  • Windows
  • macOS
  • Linux
  • Runs locally
  • Docker supported
  • Needs API key
Open websiteView on GitHub
adk-python screenshot
Popularity
21.7k Stars
GitHub stars
Recent activity
10/3/2026
Updated in the last 30 days
License
APACHE-2.0
Permissive

Why it matters

We look beyond stars: what problem it solves, whether it creates real utility, and what makes its approach worth noticing.

Last 90 days

Problem

Addresses the lack of standardized code control when building complex AI agents and workflows.

Practical value

Provides a graph-based runtime, task API, local CLI, and web UI for building and evaluating agent systems.

Innovation / differentiation

Brings traditional software engineering principles and a graph-structured execution engine to agent workflow orchestration.

Leverage potential

Integrates directly with Google platforms like Cloud Run and Vertex AI while remaining compatible with multiple models and tools.

Why now

Arrives as multi-agent systems rapidly transition from experiments to production architectures.

Community activity

In the last 90 days there were 360 new issues and 678 pull requests; the bounded issue/PR samples include 81 issue authors and 47 PR contributors, with at least 21 releases.

Maintainer responsiveness

The 100-issue window sample had a 88% close rate, and the 98-pull-request sample had a 12% merge rate. Maintainer-comment observations covered only 23% of that issue sample, so response rate and first-response speed are not reported.

47 contributors in PR sampleAt least 21 releasesPR merge rate 12% · 98 sampled in window

Key highlights

  • Graph-based Workflow Runtime with routing, parallel branches, loops, retries, and state
  • Compose single- and multi-agent systems with task delegation
  • Use functions, OpenAPI definitions, MCP servers, and other tool sources

Quick start

How it is installed, how hard it is, and where to start.

Where it runs

Runs locally

Difficulty

Medium — some setup needed

Docker supportedRuns locally
  1. 01Install the stable release in Python 3.10+ with pip install google-adk.
  2. 02Define the model, instructions, and tools with Agent, then connect nodes with Workflow when orchestration is needed.
  3. 03Use adk run or adk web to test the agent during development, and run evaluation sets with adk eval.
  4. 04For projects upgrading from 1.x, review the 2.0 breaking changes to APIs, events, and session schemas before migrating.

Best for

  • Developers who want to try it on their machine
  • People who prefer Docker deploys

More about it

Agent Development Kit (ADK) is Google’s open-source Python framework for agent development. Version 2.0 introduces a graph-based Workflow Runtime that composes agents, tasks, routing, fan-out/fan-in, loops, retries, state, and human confirmation into testable execution flows.

Agents and workflows can be defined in Python, while Agent Config offers a configuration-driven option. Tools can come from functions, OpenAPI specifications, MCP servers, and other integrations. During development, adk run provides a CLI workflow, adk web opens the development UI, and adk eval runs evaluations.

Migration needs attention: ADK 2.0 contains breaking changes to the agent API, event model, and session schema. Sessions from 2.0 are readable by ADK 1.28+ with extra fields ignored, but they are incompatible with older 1.x releases.

Sources

Each field shows its status and source — expand to review.

12 · Expand

  • capability tags

    Verified

    ai_coding, testing

    Source: manual_curated · Manually curated from the verified project README; no AI draft. · 8/17/2026

  • Latest release

    Verified

    v2.11.0

    Source: GitHub API · latest_release=v2.11.0 · 10/3/2026

  • License

    Verified

    Apache-2.0

    Source: GitHub API · license.spdx_id=Apache-2.0 · 10/3/2026

  • needs api key

    Verified

    Yes

    Source: manual_curated · Manually curated from the verified project README; no AI draft. · 8/17/2026

  • One-liner

    Verified

    {"en":"Google’s open-source Python framework for defining, orchestrating, evaluating, and deploying single- and multi-agent workflows.","zh":"Google 开源的 Python Agent 开发框架，用于定义、编排、评估并部署单 Agent 与多 Agent 工作流。"}

    Source: manual_curated · Manually curated from the verified project README; no AI draft. · 8/17/2026

  • Platforms

    Verified

    windows, macos, linux

    Source: manual_curated · Manually curated from the verified project README; no AI draft. · 8/17/2026

  • Category hint

    Inferred from materials

    ai-apps

    Source: Project README · hint=ai-apps · 8/13/2026

  • product forms

    Verified

    cli, api_sdk, library_framework

    Source: manual_curated · Manually curated from the verified project README; no AI draft. · 8/17/2026

  • role tags

    Verified

    backend, fullstack

    Source: manual_curated · Manually curated from the verified project README; no AI draft. · 8/17/2026

  • supports docker

    Verified

    Yes

    Source: Repository file · dockerfile=true; compose=false · 10/3/2026

  • supports local

    Verified

    Yes

    Source: manual_curated · Manually curated from the verified project README; no AI draft. · 8/17/2026

  • supports self host

    Verified

    No

    Source: manual_curated · Manually curated from the verified project README; no AI draft. · 8/17/2026

Related projects

Other verified projects matched by category, capabilities, and intended roles.

mirrord

A developer tool that routes a local process through a Kubernetes pod so code can use live cluster traffic, files, and environment variables without deploying first.

Same Developer tools categoryBoth support AI coding

claude-code

An agentic coding tool for the terminal that understands codebases, handles routine development tasks, explains code, and works with Git workflows.

Same Developer tools categoryBoth support AI coding

spec-kit

An open-source CLI toolkit for Spec-Driven Development that guides AI coding agents through structured specification, planning, and code generation steps.

Same Developer tools categoryBoth support AI coding

codex

OpenAI’s coding agent that runs locally, with terminal, IDE, and desktop-app entry points.

Same Developer tools categoryBoth support AI coding

puppeteer

A JavaScript library that provides a high-level API to control Chrome or Firefox over DevTools Protocol or WebDriver BiDi.

Same Developer tools categoryBoth support Testing

CLI-Anything

A toolkit for generating and managing agent-callable CLIs for existing software, with CLI-Hub, generation workflows, and skill/plugin integrations.

Same Developer tools categoryBoth support AI coding