AI Application (previously Agent Builder) Agent deployment

[andrespp123]

Hello, I have developed a conversational agent with AI Application that is grounded on a website. It responds to queries properly, but I want the agent to be part of a bigger framework, meaning I will have a second langgraph agent that would use this grounded agent to get factual responses.   

Now my question is, how can I access the grounded AI Application programmatically for the langgraph agent to query? I can't find the information in the python client documentation. Any ideas? 

[ibaui]

Hi @andrespp123,

Welcome to Google Cloud Community!

Here are some potential solutions that might address your issue:

• Wrap it in a Function or API: Create a function or a simple API endpoint that takes a question as input and returns the grounded AI's answer.
• Call the Function/API from LangGraph: In your LangGraph agent's code, call this function/API whenever you need a factual answer from the grounded source. The LangGraph agent passes the query, gets the response, and then uses that response in its overall workflow.

Was this helpful? If so, please accept this answer as “Solution”. If you need additional assistance, reply here within 2 business days and I’ll be happy to help.