From efacc7daed9600cbecd0550db161c93ac342080a Mon Sep 17 00:00:00 2001 From: adk-bot Date: Mon, 26 Jan 2026 22:25:11 +0000 Subject: [PATCH] Update ADK doc according to issue #1173 - 5 --- docs/deploy/agent-engine/deploy.md | 26 ++++++++++++++++++++++++-- 1 file changed, 24 insertions(+), 2 deletions(-) diff --git a/docs/deploy/agent-engine/deploy.md b/docs/deploy/agent-engine/deploy.md index 12b1c0803..b3c254158 100644 --- a/docs/deploy/agent-engine/deploy.md +++ b/docs/deploy/agent-engine/deploy.md @@ -1,7 +1,7 @@ # Deploy to Vertex AI Agent Engine -
- Supported in ADKPython +
+ Supported in ADKPython
This deployment procedure describes how to perform a standard deployment of @@ -146,6 +146,28 @@ For `region`, you can find a list of the supported regions on the To learn about the CLI options for the `adk deploy agent_engine` command, see the [ADK CLI Reference](https://google.github.io/adk-docs/api-reference/cli/cli.html#adk-deploy-agent-engine). +### Deploy with an API key (Express Mode) + +If you don't have a Google Cloud project set up, you can still deploy your +agent to Agent Engine using **Express Mode**, which authenticates using an +API key. This is a quicker way to get started if you don't need the full +capabilities of a Google Cloud project. + +To deploy in Express Mode, use the `--api_key` option with the +`adk deploy agent_engine` command: + +```shell +API_KEY=my-api-key + +adk deploy agent_engine \ + --api_key=$API_KEY \ + --display_name="My First Agent" \ + multi_tool_agent +``` + +When you use the `--api_key` option, the ADK initializes the Vertex AI SDK +with your API key instead of using project and location credentials. + ### Deploy command output Once successfully deployed, you should see the following output: