YAML Configuration Support and Declarative Execution in Agent Framework #1516
Unanswered
rustem-devletov
asked this question in
Q&A
Replies: 1 comment
-
|
Adding @markwallace-microsoft to help. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
I’ve been exploring the Microsoft Agent Framework and noticed that, unlike Semantic Kernel, it appears to be primarily code-first, meaning that agents, tools, and workflows are defined programmatically via Microsoft.Extensions.AI abstractions rather than declaratively.
My question is about YAML support and declarative configurations:
In Semantic Kernel, YAML (or prompt templates) can be directly mounted and executed as “skills” or “plugins.” In Agent Framework, it seems YAML could still be used for configuration, but would need to be manually parsed and mapped to ChatClientAgentOptions, ChatOptions, etc. (for example, using YamlDotNet).
Could you please clarify whether MAF currently supports any form of declarative execution (e.g., auto-loading or binding YAML-defined agents/workflows)? If not, is there a planned roadmap or recommended pattern for integrating YAML-based configurations to instantiate and wire agents dynamically at runtime? Are there examples or guidelines for best practices in using YAML (or similar) alongside the Agent Framework for configuration-driven agent definitions?
Thanks in advance!
Rustem.
Beta Was this translation helpful? Give feedback.
All reactions