repowiserepowise

MCP server

google / adk-python

An open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.

★ 22,230stars on GitHubPythonApache-2.0

This connects Repowise to google/adk-python's code. It does not install adk-python.

Repowise maps public repos from their source. Not affiliated with google. What is Repowise?

github.com/google/adk-pythonIndexed at f4fd7d5

What is adk-python?

adk-python is an open-source, code-first Python framework for building, evaluating, and deploying AI agents. Developers define an agent in Python code by giving it a name, a model, and instructions, and tools and behavior can be added to it. Agents can be composed into graph-based workflows that handle routing, loops, retries, and human-in-the-loop steps. The project is written in Python and installed from PyPI as google-adk. To install it, run pip install google-adk and follow the steps below. Its code health is 6.7 out of 10 (Fair), ranked #157 of 4,000+ indexed repos. The last commit to adk-python was on Aug 11, 2026, and 100 people have contributed.

As of commit f4fd7d5, indexed Aug 11, 2026. What it is comes from a model reading the README; the rest is measured from the source and git history, not from the maintainer.

For agentsadk-python.md

Repo facts: type, license, health, rank, contributors
adk-python facts
TypeMCP server
LicenseApache-2.0
Code health6.7 / 10 (Fair)
Health rank#157 of 4,000+ indexed repos
Last commitAug 11, 2026
Contributors100
Bus factor1
Size2,369 files, 418,576 lines

Install adk-python

Commands copied from the README

  1. 1Install the package from PyPI.
    pip install google-adk
  2. 2Run an agent in the interactive CLI.
    adk run path/to/my_agent
  3. 3Start the web UI for an agents directory.
    adk web path/to/agents_dir

Releases on GitHub (latest v2.14.1, Nov 12, 2026)

Full install notes on GitHub

Explore adk-python

Ask about this code in your agent

This connects Repowise to adk-python's code. It does not install adk-python. To install adk-python, use the install steps.

Useful if you want to change or fork adk-python: your agent can ask how it is put together and get answers that cite its files.

Ask about adk-python

Answers come from adk-python's README and source, with the files they cite. Two free answers a day without an account.

Want this for your own codebase? Map your repo free

About this page

Get a page like this for your own codebase

Repowise built this map of google/adk-python from its source and git history. It parses every symbol, draws the dependency graph, scores each file's health from complexity, duplication, coverage and churn, finds hotspots and owners in the history, and writes documentation for every module. Your coding agent reads the same index over MCP, so its answers cite real files.

Free for 2 public repos, with a page like this one. No card. See how Repowise works

Prefer local? pip install repowise && repowise init indexes your own checkout, no account needed. Star Repowise on GitHub

google/adk-python FAQ

What is adk-python?

adk-python is an open-source, code-first Python framework for building, evaluating, and deploying AI agents. It provides an Agent class for defining instructions, tools, and behavior, and a Workflow class for graph-based orchestration.

How does adk-python work?

adk-python lets you define an agent in Python with a name, a model, and an instruction, then run it with adk run or adk web. Workflows compose agents and tasks into graph-based flows with routing, loops, and retry.

How do I install adk-python?

adk-python is installed with pip install google-adk. Python 3.10 or newer is required. See the install steps

What language is adk-python written in?

adk-python is written in Python and is published on PyPI as google-adk.

How is the adk-python code organized?

Package code lives in src and unit tests live in tests. Runnable example agents are in contributing/samples, and task-oriented walkthroughs are in docs/guides.

Who is adk-python for?

adk-python is for developers who build, evaluate, and deploy AI agents in Python and want flexibility and control over them.

What license does adk-python use?

adk-python uses the Apache-2.0 license.

Code health

6.7out of 10Fair

Hotspots average 4.8, the place fixes pay off most.

On the leaderboards:#5 of 5 Agent frameworks#43 of 67 Python repos

Key features of adk-python

  • A Workflow runtime that composes deterministic execution flows from graph-based nodes, with routing, fan-out and fan-in, loops, retry, state management, and nested workflows.
  • Human-in-the-loop steps and dynamic nodes that can be used inside workflows.
  • A Task API for structured agent-to-agent delegation, with multi-turn task mode and single-turn controlled output.
  • An adk run command for an interactive CLI and an adk web command for a web UI.

How google/adk-python works

Agents are defined in Python with the Agent class from google.adk, which takes a name, a model, and an instruction. The Workflow runtime orchestrates agents and tasks as a graph, supporting routing, fan-out and fan-in, loops, and nested workflows. The src directory holds the package code, including the google/adk/plugins module and its bigquery_agent_analytics_plugin.py file. Agents are run through the adk command, where adk run starts an interactive CLI and adk web starts a web UI for an agents directory. The docs/guides directory covers agents, tools, events, plugins, and workflows, and contributing/samples contains runnable example agents such as contributing/samples/adk_team/adk_answering_agent/main.py. Unit tests are kept under tests/unittests.

The codebase has 2,369 files and 24,391 symbols in 5 modules, led by Python, JavaScript and Terraform. Code health is 6.7 out of 10, rated fair.