Use when building AI agents with Google's Agent Development Kit (ADK). Triggers include creating ADK agents, Gemini-powered agents, multi-agent workflows, MCP integration with ADK, agent evaluation, workflow agents (Sequential/Parallel/Loop), Vertex AI deployment, or comparing OpenAI SDK with ADK. NOT when using other frameworks (OpenAI SDK, LangChain, CrewAI) or general LLM API calls.
SkillScore100Maintained
Automatic
◆AI Engineer pack
Last verified 8/18/2026
What it changes
Passed 22 of 22 test cases with the skill, and 8 of 22 without it.
Without skill36%
With skill100%
Checks it passed that the model alone missed
Includes --service_name flag
Uses adk deploy cloud_run command
Wraps child agents using AgentTool
and 18 more in the full report
14 fixed8 passed either way0 still failing0 broke
Pass-rate lift
+64 pp
Tokens
+140%
Agent turns
0%
Response time
-50%
Model gemini-3.6-flashMeasured 8/18/2026v122 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 14
Prompt
Provide the CLI command to deploy a local Google ADK agent folder named ./my_agent to Vertex AI Agent Engine with staging bucket gs://my-bucket in us-central1.
What was checked, without and with the skill
Checked
Without skill
With skill
Uses adk deploy agent_engine command
Passed
Passed
Includes --staging_bucket flag with the GCS bucket path
Missed
Passed
case-15 · gemini-3.6-flash · 8/18/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md645 linessingle file
Building with Google ADK
Build production-grade AI agents using Google's Agent Development Kit.
Quick Start
# Install ADK
pip install google-adk
# Or with uv (recommended)
uv add google-adk
# Create new agent project
adk create my_agent
# Run locally
adk run my_agent
# Web UI for debugging
adk web