# AI use cases for Adk Python (google/adk-python) Repo: [google/adk-python](https://github.com/google/adk-python) | Use cases: 3 | Page: https://dedupedai.com/use-cases/google-adk-python/ Agent Development Kit (ADK) is Google's open-source, code-first Python framework for building, evaluating and deploying AI agents. Agents and graph-based Workflows are defined in Python, with routing, loops, retries, human-in-the-loop and a Task API for agent-to-agent delegation, plus pre-built, OpenAPI and MCP tools. It ships a CLI with an interactive runner, a development web UI, evaluation and one-command deploys to Docker or Cloud Run. It is optimised for Gemini but model-agnostic. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Build an agent and test it in ADK's web UI: https://dedupedai.com/use-cases/google-adk-python/build-and-test-an-agent-in-the-adk-web-ui/ - Chain agents into a deterministic workflow: https://dedupedai.com/use-cases/google-adk-python/chain-agents-in-an-adk-workflow/ - Evaluate an agent and deploy it to Cloud Run: https://dedupedai.com/use-cases/google-adk-python/evaluate-and-deploy-an-adk-agent/ ## How to build an agent and test it in ADK's web UI with Adk Python Repo: [google/adk-python](https://github.com/google/adk-python) | Difficulty: Beginner | Time: 20 minutes | Basis: From the README Define an agent in a few lines of Python and chat with it in the built-in development UI. **Who it is for:** A developer prototyping an assistant who wants to see tool calls and responses before writing an app around it. **The problem:** Testing an agent from print statements makes it hard to see what it did and why. **The outcome:** A working root_agent you can run from the terminal or debug in the ADK web UI. ### Steps 1. Install with `pip install google-adk` (Python 3.11+) 2. Create an Agent with a name, model and instruction, as in the Quick Start, and assign it to root_agent 3. Chat with it in the terminal with `adk run path/to/my_agent` 4. Open the development UI with `adk web path/to/agents_dir` to test, debug and showcase it 5. Add optional integrations with `pip install "google-adk[extensions]"` when you need them ### Prompt to paste into your agent ```text Using google/adk-python, create an agent folder with a root_agent that , give it , and run it with `adk web path/to/agents_dir`. Show me a test conversation and the tool calls it made. ``` ### Caveat Releases land roughly every two weeks and this is ADK 2.0, so older tutorials may not match; the README also offers constraints files for safer installs. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/google-adk-python/build-and-test-an-agent-in-the-adk-web-ui/ ## How to chain agents into a deterministic workflow with Adk Python Repo: [google/adk-python](https://github.com/google/adk-python) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Use ADK's Workflow runtime to pass work between specialised agents in a fixed graph. **Who it is for:** A builder whose single agent keeps skipping steps in a multi-stage process. **The problem:** Leaving every step to one model's judgment makes multi-stage tasks unpredictable. **The outcome:** A Workflow whose edges run one agent after another, with routing, loops and retries available. ### Steps 1. Install with `pip install google-adk` 2. Define each step as an Agent with its own instruction 3. Create a Workflow with edges such as START to the first agent to the second, as in the README example 4. Add routing, fan-out and fan-in, loops, retry or human-in-the-loop steps as the Workflow Runtime supports 5. Run it with adk run or adk web and check each node's output ### Prompt to paste into your agent ```text With google/adk-python, build a Workflow where does and passes its output to for . Add a retry on and a human-in-the-loop check before , then run it in adk web. ``` ### Caveat Only the simplest Workflow is shown in the README; routing and loops are in the guides under docs/guides. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/google-adk-python/chain-agents-in-an-adk-workflow/ ## How to evaluate an agent and deploy it to Cloud Run with Adk Python Repo: [google/adk-python](https://github.com/google/adk-python) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Run ADK's evaluation against a test set, then ship the agent in a container or on Cloud Run with one command. **Who it is for:** A team moving an agent from prototype to a shared service that needs evidence it still behaves. **The problem:** Agents regress silently when prompts or tools change, and deploying them needs custom container work. **The outcome:** An evaluated agent running locally in Docker or on Cloud Run, with the web UI included. ### Steps 1. Try the bundled evaluation sample from the repo root with adk eval, pointing it at the home_automation_agent, its evalset and its eval config 2. Write your own evalset for your agent and rerun adk eval after each change 3. Deploy locally in a container with `adk deploy docker --with_ui` 4. Deploy to Google Cloud with `adk deploy cloud_run --with_ui` 5. Set environment variables on the command or in a .env file, for example `adk deploy cloud_run --with_ui --env GOOGLE_GENAI_USE_ENTERPRISE=1` ### Prompt to paste into your agent ```text Using google/adk-python, write an evalset for my agent covering , run adk eval and report failures. If it passes, deploy it with `adk deploy cloud_run --with_ui` and give me the URL. ``` ### Caveat The README shows the deploy commands but not the Google Cloud setup they need; follow the docs at adk.dev first. Risk, credentials: Deploying to Cloud Run uses your Google Cloud credentials and can incur charges; the --with_ui flag also deploys the web UI, so restrict who can reach it. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/google-adk-python/evaluate-and-deploy-an-adk-agent/