In short:
ak-google-adk-pythonbuilds agents with Google ADK for Python: a single agent with tools, a chain of agents, an MCP server as a tool, a remote agent over A2A, and deployment when you ask for it. It writes code for thegoogle-adkversion your project already has and always adds a task or eval to check the result. It does not upgrade the library just to match a sample, and it has no fixed next skill.
Disclosure: I maintain AgentKit and have an affiliate relationship with it: the AgentKit link in this post is an affiliate link; buying through it gets you 30% off.
Summary
Use it when you want to build or extend an agent written with Google ADK Python, from a single agent all the way to Cloud Run, Vertex AI or GKE. You do not need it when the project does not use ADK, or when you only have a general question about agents and no code to write.
flowchart LR
A["ADK agent idea"] --> B["ak-google-adk-python"]
B --> C["Agent code for the installed SDK"]
B --> D["Task or eval to check it"]
C --> E["Test run, deploy when you ask"]Four things the skill always holds to when it writes an agent:
| Concept | Question | Example |
|---|---|---|
| root_agent or app entry point | Where does ADK look for your agent? | __init__.py has from . import agent, and agent.py declares root_agent = Agent(...). For plugins or event compaction in long sessions, declare app = App(...). |
| Follow the installed SDK | Does the docs example run on your project's google-adk version? | The sample uses a preview model your current model list does not have. The skill keeps your model and adapts the code to the SDK, instead of upgrading google-adk to fit the sample. |
| Pick the right agent type | Is the flow fixed, or does the model decide whom to call? | A fixed order uses SequentialAgent. A coordinator routing questions to a "billing" or "tech" agent uses Agent with sub_agents. Several review rounds use LoopAgent with max_iterations. |
| Check by observation | What evidence shows the agent works? | An evalset with "2+2?" expecting "4". Each run records the model, errors, time, and cost when telemetry exists, otherwise cost is marked unknown. |
Trade-off: sticking to the installed SDK sometimes means adapting a sample instead of copying it as is. The exception is losing the session when an MCP tool errors: current ADK already keeps the session, so upgrading google-adk is the right fix there.
Sample prompt (single agent with tools):
/ak-google-adk-python Create an order support agent for <project name>.
Tools: look up order status by order ID, compute shipping fee by province.
Constraints: keep the installed google-adk version and the model the project already uses.
Done when: there is an evalset with a few sample questions and the agent answers them correctly.This skill has no flags.
Below is the detailed version: the run flow, 5 situations with prompts, the change history by ak release, and common pitfalls. The detailed version and the interactive page are for subscribers. Your agent can also search them through the portal's MCP.
Opens when your account has subscriber access
This guide is for subscribers
Once your account has access, the full piece, the interactive visual and the agent call all open.
- Sign in or create a modernEngineer accountReading access belongs to your account email.
- Choose a plan on the pricing page1, 3 or 12 months, paid by VietQR bank transfer. Access opens about a minute after the money arrives.
- Or redeem a Facebook subscriber codePeople supporting the page on Facebook get an activation code by message each month.
Already a subscriber? Sign in to read.
Use this piece in your agent
Get an API keyCopy the text below and paste it into Claude Code, Codex or whichever agent you use. The agent reads the piece and checks it against the repo you have open.
Read "ak-google-adk-python: build Google ADK agents on your installed SDK" on modernEngineer and apply it to this repo.
1. Call the content_get tool of the modernengineer MCP server with locale "en", slug "ak-google-adk-python". If the modernengineer MCP server is not connected, stop and walk me through connecting it from https://modernengineer.dev/en/mcp.
2. Summarize the main steps of the piece and point out which apply to this repo.
3. Propose concrete changes and ask me before editing any file.Connect MCP in Claude Code (once)
claude mcp add --transport http modernengineer \
https://api.modernengineer.dev/mcp \
--header "Authorization: Bearer <API key>"Reading the full piece over MCP needs an API key from a subscriber account or one that bought this guide. Without access, the agent only gets the opening.
- Version
- v1.0.0
- Updated
- 22/10/2026
- Licence
- All rights reserved
- Slug
- ak-google-adk-python
