Structured Output from Flow Agent Actions in Winter '26
Summary
The Winter '26 release introduces structured output support for AI Agent Actions in Salesforce Flows, enabling low-code automation to convert unstructured LLM responses into typed Apex objects. This lets admins and developers define fields for AI-generated data directly within Flows without custom Apex development. The blog demonstrates a simple screen flow example calculating opportunity totals using Copilot for Salesforce as an AI Agent, clarifying that the platform manages the structured output creation. This feature empowers Salesforce teams to enhance automation with AI-driven data extraction while maintaining low-code simplicity.
Takeaways
- AI Agent Actions can return structured output accessible as dynamic Apex classes.
- Low-code admins can define output fields without writing custom Apex code.
- Salesforce platform converts LLM text responses into usable complex data types.
- Use AI Agent Actions in Flows to augment automation with AI-derived insights.
- Avoid unnecessary LLM calls for simple calculations better handled in Flow.
Image created by ChatGPT 5o based on a prompt by Bob Buzzard Introduction When preparing for the Credera Winter '26 release webinar , the release notes for this feature gave me significant pause for thought. Not because it was an awesome change that I'd been waiting ages for, nor that it was something out of left field that I couldn't wait to try. Instead it was because I didn't understand how it worked. The release notes talked about custom agent actions returning specific fields, so did that mean it was the action itself that returned complex data types? There was only one way to find out. Giving it a Go Once I'd waited for my Agentforce developer edition to be upgraded to Winter '26 I was able to try out this new functionality. In order to understand how much effort I had to put in around my actions, I started off putting in zero effort.