diff --git a/docs/assets/adk-web-dev-ui-audio.png b/docs/assets/adk-web-dev-ui-audio.png new file mode 100644 index 0000000000..fdc9d0a09f Binary files /dev/null and b/docs/assets/adk-web-dev-ui-audio.png differ diff --git a/docs/assets/adk-web-dev-ui-chat.png b/docs/assets/adk-web-dev-ui-chat.png new file mode 100644 index 0000000000..5fd4faa555 Binary files /dev/null and b/docs/assets/adk-web-dev-ui-chat.png differ diff --git a/docs/assets/adk-web-dev-ui-function-call.png b/docs/assets/adk-web-dev-ui-function-call.png new file mode 100644 index 0000000000..61d3afcea8 Binary files /dev/null and b/docs/assets/adk-web-dev-ui-function-call.png differ diff --git a/docs/assets/agent-interaction.png b/docs/assets/agent-interaction.png new file mode 100644 index 0000000000..a03e5f4200 Binary files /dev/null and b/docs/assets/agent-interaction.png differ diff --git a/docs/deploy/agent-engine.md b/docs/deploy/agent-engine.md index 47cc8ac80a..166196aef8 100644 --- a/docs/deploy/agent-engine.md +++ b/docs/deploy/agent-engine.md @@ -64,7 +64,7 @@ Use `reasoning_engines.AdkApp()` to wrap your agent to make it deployable to Age from vertexai.preview import reasoning_engines app = reasoning_engines.AdkApp( - agent_engine=root_agent, + agent=root_agent, enable_tracing=True, ) ``` diff --git a/docs/get-started/quickstart.md b/docs/get-started/quickstart.md index 2fea90811f..b06058533d 100644 --- a/docs/get-started/quickstart.md +++ b/docs/get-started/quickstart.md @@ -113,15 +113,9 @@ parent_folder/ <-- navigate to this directory .env ``` -=== "adk run" - - Run the following command, to chat with your Google Search agent. - - ``` - adk run multi_tool_agent - ``` - To exit, use Cmd/Ctrl+C. +There are multiple ways to interact with your agent: +![agent-interaction.png](../assets/agent-interaction.png) === "adk web (dev UI)" Run the following command to launch the **dev UI**. @@ -129,16 +123,40 @@ parent_folder/ <-- navigate to this directory adk web ``` - Open the URL provided (usually `http://localhost:8000` or - `http://127.0.0.1:8000`) **directly in your browser**. This connection stays - entirely on your local machine. Select `multi_tool_agent` and interact with the agent. + **Step 1:** Open the URL provided (usually `http://localhost:8000` or `http://127.0.0.1:8000`) directly in your browser. + + **Step 2.** In the top-left corner of the UI, you can select your agent in the dropdown. Select "my_sample_agent". - !!!note + !!!note "Troubleshooting" If you do not see "my_sample_agent" in the dropdown menu, make sure you are running `adk web` in the **parent folder** of your agent folder (i.e. the parent folder of my_sample_agent). - 1. In the top-left corner of the UI, you can select your agent in the dropdown. Select "my_sample_agent". - 2. Now you can chat with your agent. + **Step 3.** Now you can chat with your agent using the textbox: + + ![adk-web-dev-ui-chat.png](../assets/adk-web-dev-ui-chat.png) + + **Step 4.** You can also inspect individual function calls, responses and model responses by clicking on the actions: + + ![adk-web-dev-ui-function-call.png](../assets/adk-web-dev-ui-function-call.png) + + **Step 5.** You can also enable your microphone and talk to your agent: + + ![adk-web-dev-ui-audio.png](../assets/adk-web-dev-ui-audio.png) + + !!!note "Model support" + + Currently only `gemini-2.0-flash-exp` supports talking to your agent via audio/video. For more information, see the [documentation on the Live API on Vertex AI](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal-live-api) + + +=== "adk run" + + Run the following command, to chat with your Google Search agent. + + ``` + adk run multi_tool_agent + ``` + + To exit, use Cmd/Ctrl+C. ### 📝 Example prompts to try diff --git a/examples/python/snippets/get-started/google_search_agent/agent.py b/examples/python/snippets/get-started/google_search_agent/agent.py index 241f521486..171d2b7d1b 100644 --- a/examples/python/snippets/get-started/google_search_agent/agent.py +++ b/examples/python/snippets/get-started/google_search_agent/agent.py @@ -3,7 +3,7 @@ root_agent = Agent( # A unique name for the agent. - name="basic_search_agent", + name="google_search_agent", # The Large Language Model (LLM) that agent will use. model="gemini-2.0-flash-exp", # A short description of the agent's purpose.