adk-python

View on GitHub

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

★ 22K4,059 forksPythonApache-2.0google

Google's code-first Python framework for building, orchestrating, evaluating and deploying AI agents. Offers graph workflow runtime, multi-agent task delegation, MCP/OpenAPI tools, HITL confirmation, a dev web UI and Cloud Run / Vertex AI deployment. Model-agnostic but optimized for Gemini.

Use Cases

Build code-first LLM agents in PythonCompose multi-agent hierarchies and delegationGraph-based workflow orchestration with loops and routingAdd custom functions, OpenAPI or MCP tools to agentsHuman-in-the-loop tool confirmation flowsEvaluate agent quality with eval setsDebug and demo agents in the built-in ADK Web dev UIDeploy agents to Cloud Run or Vertex AI Agent EngineA2A agent-to-agent communicationGenerate agents from config without code

Built With

Language: Python
Frameworks: Google GenAI SDK · FastAPI · Starlette · Pydantic · OpenTelemetry · aiosqlite · uvicorn · a2a-sdk · MCP · Graphviz · LiteLLM

adk-python

Use Cases

Built With

Tags