adk-python

Name: adk-python
Author: google

bygoogle· 1 MCP server

Stars

21k

Forks

3.8k

How to install adk-python?

google/adk-python

$git clone https://github.com/google/adk-python

Installs into the current project.

›Prefer a prompt? Paste this to your agent

Install & use

Install adk-python by running `git clone https://github.com/google/adk-python`, then use it for the current task and follow its documentation at https://github.com/google/adk-python.

Files · 1

View on GitHub

README.md

1	# Agent Development Kit (ADK) 2.0
2
3	[![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
4	[![PyPI version](https://img.shields.io/pypi/v/google-adk.svg)](https://pypi.org/project/google-adk/)
5	[![Python versions](https://img.shields.io/pypi/pyversions/google-adk.svg)](https://pypi.org/project/google-adk/)
6	[![PyPI downloads](https://static.pepy.tech/badge/google-adk/month)](https://pepy.tech/project/google-adk)
7	[![Unit Tests](https://github.com/google/adk-python/actions/workflows/python-unit-tests.yml/badge.svg)](https://github.com/google/adk-python/actions/workflows/python-unit-tests.yml)
8	[![Docs](https://img.shields.io/badge/docs-latest-blue.svg)](https://google.github.io/adk-docs/)
9
10	<h2 align="center">
11	<img src="https://raw.githubusercontent.com/google/adk-python/main/assets/agent-development-kit.png" width="256"/>
12	</h2>
13	<h3 align="center">
14	An open-source, code-first Python framework for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
15	</h3>
16	<h3 align="center">
17	Important Links:
18	<a href="https://google.github.io/adk-docs/">Docs</a>,
19	<a href="https://github.com/google/adk-samples">Samples</a> &
20	<a href="https://github.com/google/adk-web">ADK Web</a>.
21	</h3>
22
23	______________________________________________________________________
24
25	> ⚠️ BREAKING CHANGES FROM 1.x
26	>
27	> This release includes breaking changes to the agent API, event model, and
28	> session schema. **Sessions generated by ADK 2.0 are readable by ADK 1.28+
29	> (extra fields will be ignored), but are incompatible with older 1.x versions.**
30
31	______________________________________________________________________
32
33	## 🔥 What's New in 2.0
34
35	- Workflow Runtime: A graph-based execution engine for composing
36	deterministic execution flows for agentic apps, with support for routing,
37	fan-out/fan-in, loops, retry, state management, dynamic nodes,
38	human-in-the-loop, and nested workflows.
39
40	- Task API: Structured agent-to-agent delegation with multi-turn task
41	mode, single-turn controlled output, mixed delegation patterns,
42	human-in-the-loop, and task agents as workflow nodes.
43
44	## 🚀 Installation
45
46	```bash
47	pip install google-adk
48	```
49
50	Requirements: Python 3.10+.
51
52	To install optional integrations, you can use the following command:
53
54	```bash
55	pip install "google-adk[extensions]"
56	```
57
58	The release cadence is roughly bi-weekly.
59
60	## Quick Start
61
62	> Beginner Note: ADK applications are built using two main classes:
63	> `Agent` (defines an AI's instructions, tools, and behavior) and
64	> `Workflow` (orchestrates agents and tasks in a graph-based flow).
65
66	### Agent
67
68	```python
69	from google.adk import Agent
70
71	root_agent = Agent(
72	name="greeting_agent",
73	model="gemini-2.5-flash",
74	instruction="You are a helpful assistant. Greet the user warmly.",
75	)
76	```
77
78	### Workflow
79
80	```python
81	from google.adk import Agent, Workflow
82
83	generate_fruit_agent = Agent(
84	name="generate_fruit_agent",
85	instruction="Return the name of a random fruit. Return only the name.",
86	)
87
88	generate_benefit_agent = Agent(
89	name="generate_benefit_agent",
90	instruction="Tell me a health benefit about the specified fruit.",
91	)
92
93	root_agent = Workflow(
94	name="root_agent",
95	edges=[("START", generate_fruit_agent, generate_benefit_agent)],
96	)
97	```
98
99	### Run Locally
100
101	```bash
102	# Interactive CLI
103	adk run path/to/my_agent
104
105	# Web UI (supports multi-agent directories or pointing directly to a single agent folder)
106	adk web path/to/agents_dir
107	```
108
109	## 📚 Documentation
110
111	- Getting Started: https://google.github.io/adk-docs/
112	- Samples: See `contributing/workflow_samples/` and
113	`contributing/task_samples/` for workflow and task API examples.
114
115	## 🤝 Contributing
116
117	See [CONTRIBUTING.md](CONTRIBUTING.md) for details.
118
119	## 📄 License
120
121	This project is licensed under the Apache 2.0 License — see the
122	[LICENSE](LICENSE) file for details.

6 picks

Type

adk-python

bygoogle· 1 MCP server

Stars

21k

Forks

3.8k

How to install adk-python?

google/adk-python

$git clone https://github.com/google/adk-python

Installs into the current project.

›Prefer a prompt? Paste this to your agent

Install & use

Install adk-python by running `git clone https://github.com/google/adk-python`, then use it for the current task and follow its documentation at https://github.com/google/adk-python.

Files · 1

View on GitHub

README.md

1	# Agent Development Kit (ADK) 2.0
2
3	[![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
4	[![PyPI version](https://img.shields.io/pypi/v/google-adk.svg)](https://pypi.org/project/google-adk/)
5	[![Python versions](https://img.shields.io/pypi/pyversions/google-adk.svg)](https://pypi.org/project/google-adk/)
6	[![PyPI downloads](https://static.pepy.tech/badge/google-adk/month)](https://pepy.tech/project/google-adk)
7	[![Unit Tests](https://github.com/google/adk-python/actions/workflows/python-unit-tests.yml/badge.svg)](https://github.com/google/adk-python/actions/workflows/python-unit-tests.yml)
8	[![Docs](https://img.shields.io/badge/docs-latest-blue.svg)](https://google.github.io/adk-docs/)
9
10	<h2 align="center">
11	<img src="https://raw.githubusercontent.com/google/adk-python/main/assets/agent-development-kit.png" width="256"/>
12	</h2>
13	<h3 align="center">
14	An open-source, code-first Python framework for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
15	</h3>
16	<h3 align="center">
17	Important Links:
18	<a href="https://google.github.io/adk-docs/">Docs</a>,
19	<a href="https://github.com/google/adk-samples">Samples</a> &
20	<a href="https://github.com/google/adk-web">ADK Web</a>.
21	</h3>
22
23	______________________________________________________________________
24
25	> ⚠️ BREAKING CHANGES FROM 1.x
26	>
27	> This release includes breaking changes to the agent API, event model, and
28	> session schema. **Sessions generated by ADK 2.0 are readable by ADK 1.28+
29	> (extra fields will be ignored), but are incompatible with older 1.x versions.**
30
31	______________________________________________________________________
32
33	## 🔥 What's New in 2.0
34
35	- Workflow Runtime: A graph-based execution engine for composing
36	deterministic execution flows for agentic apps, with support for routing,
37	fan-out/fan-in, loops, retry, state management, dynamic nodes,
38	human-in-the-loop, and nested workflows.
39
40	- Task API: Structured agent-to-agent delegation with multi-turn task
41	mode, single-turn controlled output, mixed delegation patterns,
42	human-in-the-loop, and task agents as workflow nodes.
43
44	## 🚀 Installation
45
46	```bash
47	pip install google-adk
48	```
49
50	Requirements: Python 3.10+.
51
52	To install optional integrations, you can use the following command:
53
54	```bash
55	pip install "google-adk[extensions]"
56	```
57
58	The release cadence is roughly bi-weekly.
59
60	## Quick Start
61
62	> Beginner Note: ADK applications are built using two main classes:
63	> `Agent` (defines an AI's instructions, tools, and behavior) and
64	> `Workflow` (orchestrates agents and tasks in a graph-based flow).
65
66	### Agent
67
68	```python
69	from google.adk import Agent
70
71	root_agent = Agent(
72	name="greeting_agent",
73	model="gemini-2.5-flash",
74	instruction="You are a helpful assistant. Greet the user warmly.",
75	)
76	```
77
78	### Workflow
79
80	```python
81	from google.adk import Agent, Workflow
82
83	generate_fruit_agent = Agent(
84	name="generate_fruit_agent",
85	instruction="Return the name of a random fruit. Return only the name.",
86	)
87
88	generate_benefit_agent = Agent(
89	name="generate_benefit_agent",
90	instruction="Tell me a health benefit about the specified fruit.",
91	)
92
93	root_agent = Workflow(
94	name="root_agent",
95	edges=[("START", generate_fruit_agent, generate_benefit_agent)],
96	)
97	```
98
99	### Run Locally
100
101	```bash
102	# Interactive CLI
103	adk run path/to/my_agent
104
105	# Web UI (supports multi-agent directories or pointing directly to a single agent folder)
106	adk web path/to/agents_dir
107	```
108
109	## 📚 Documentation
110
111	- Getting Started: https://google.github.io/adk-docs/
112	- Samples: See `contributing/workflow_samples/` and
113	`contributing/task_samples/` for workflow and task API examples.
114
115	## 🤝 Contributing
116
117	See [CONTRIBUTING.md](CONTRIBUTING.md) for details.
118
119	## 📄 License
120
121	This project is licensed under the Apache 2.0 License — see the
122	[LICENSE](LICENSE) file for details.

6 picks

Type

adk-python

How to install adk-python?

Related

adk-python

How to install adk-python?

Related