Agent Development Kit
Google's agent framework: agents as Python code, with a workflow runtime and an evaluation harness included
What is Agent Development Kit?
You write an agent as an LlmAgent — a model, an instruction and a list of tools — then compose several into a hierarchy, or wire them into a workflow whose sequential, parallel and loop steps you control rather than leave to the model. A Runner drives each turn and keeps the session and its state, so a conversation survives across requests. It ships the parts most frameworks leave out: an evaluation harness that scores agents against saved cases, and a local web UI for stepping through a run. It is model-agnostic through LiteLLM and runs anywhere in a container, but the documentation, the tool catalogue and the managed runtime are Google's, so the further you get from Gemini and Google Cloud the more you assemble yourself.
What can you do with Agent Development Kit?
- Write the agent, not the plumbing — An LlmAgent takes a model, an instruction and Python functions as tools; type hints and docstrings become the tool description the model reads, so there is no separate schema to maintain.
- Decide the order yourself where it matters — Sequential, parallel and loop agents run their children in a fixed shape, so a pipeline that must always retrieve before it answers is expressed in code rather than hoped for in a prompt.
- Delegate between specialised agents — Agents nest into a hierarchy and hand work to each other, and the Task API makes that delegation explicit — including a step that waits for a person to approve before a tool runs.
- Score agents against saved cases — The evaluation harness replays stored sessions and grades both the final answer and the tool calls taken to reach it, so a prompt change that quietly breaks a path shows up as a number.
- Step through a run in the browser —
adk webopens a local UI listing every event of a turn — the model calls, the tool calls and the state after each — which is where an agent that loops or picks the wrong tool becomes visible. - Bring tools you already have — Alongside Google's own catalogue, ADK loads tools from an OpenAPI spec or an MCP server, and can wrap an agent built in LangChain or CrewAI as a tool of its own.
Before you choose Agent Development Kit
- Agents have been reported repeating a tool call the instruction told them to make once, and combining a plain `str` output schema with tools has looped indefinitely between the model and its tools.Reported in#3940#5054
- Version 2.0 changed the agent API, the event model and the session schema: sessions written by 2.0 cannot be read by 1.x, and the project states its release cadence is roughly every two weeks.
Star history
17 Aug to 28 Aug · +156
Frequently asked questions
Is Agent Development Kit free for commercial use?
Agent Development Kit is released under the Apache-2.0 licence — OSI-approved open source, which permits commercial use.
How can Agent Development Kit be deployed?
Agent Development Kit is available as Self-hosted / Managed cloud.
Other options in Agent Frameworks
deepseek-harnessFrameworks`dsh` is the agent harness DeepSeek AI develops itself. `npx @deepseek-ai/dsh web` brings up a Web UI on `http://127.0.0.1:3080`, and the architecture behind it is one where everything is a plugin, powered by the Cordis runtime, so it suits people who intend to extend a harness rather than only operate one. The caveat comes from the project itself: it is labelled a developer preview and warns in capitals that there will be compatibility-breaking changes, so plugins written today should budget for rework. It is also a poor fit if you want a documented scripted or embedded entry point right now, since the README covers only the `web` command and a source checkout, and links out to a development guide and architecture documentation without saying what either contains.
AutoGPTWorkflowAn agent here is a graph of blocks — one block calls a service, transforms data, prompts a model or branches — assembled on a canvas rather than written in code, then run on demand, on a schedule or from a trigger. One repository holds both the free self-hosted platform and the code behind the paid hosted one, and that is the catch: everything under the platform directory is PolyForm Shield, which permits use but forbids offering it as a competing product.
DifyWorkflowA complete platform — prompt orchestration, retrieval, agent tools and an admin UI — that non-engineers can operate once a team has set it up. The licence is a modified Apache 2.0 with additional conditions on multi-tenant hosting and branding, so it is source-available rather than open source. Read it before building a product on top.