Share via

Why is my Copilot Agent sending an "explanation_of_tool_call" message after sending sub-agent response?

John Fox 0 Reputation points
2025-12-11T17:40:18.2233333+00:00

Hey all,

We have a Copilot Agent with the following issue.

My agent (we'll call A) has no external knowledge sources, nor have I edited the topics, so I wouldn't anticipate any issues there.

A does have three sub agents (B, C, and D). A's connections to these sub agents are designed so after A determines it needs to use one of them, it should not respond (the completion setting set to "Don't respond (default)").

My issue comes from the fact that when I ask A a question in any environment (including the test window) it will determine the sub agent to answer the question (correctly), have the sub agent answer the question (which is pushed directly to the chat I am having with A, also correctly), and then push a second message formatted as follows:

{"explanation_of_tool_call": the actual explanation of what the agent did here}

This message does not appear when I ask any of B, C, or D a question. It only happens with A.

I read in another forum that bug ID CSUP-10877 was dealing with a similar issue, but that dealt with someone's specific Tool issue while mine has to deal with a sub-agent call I presume. I did hear in the comments there that there are others encountering my issue.

Any advice from the community and Microsoft would be greatly appreciated!!

John

Microsoft Copilot | Microsoft 365 Copilot | Development

2 answers

Sort by: Most helpful
Most helpful Newest Oldest
  1. Deleted

    This answer has been deleted due to a violation of our Code of Conduct. The answer was manually reported or identified through automated detection before action was taken. Please refer to our Code of Conduct for more information.

    Comments have been turned off. Learn more

  2. Karan Shewale 2,385 Reputation points Microsoft External Staff
    2025-12-12T08:33:18.6533333+00:00

    Hi John Fox,

    This happens because Agent A is still using an older system for handling sub-agents. Even though Agent A is set to “Don’t respond”, the platform sometimes sends an extra internal message called explanation_of_tool_call after the sub-agent replies.

    Key points

    The extra message is created by the system, not by sub-agents B, C, or D.

    It usually shows up only when the parent agent doesn’t have any topics or knowledge sources and is relying fully on sub-agents.

    Microsoft already knows about this and is updating the runtime so these internal messages won’t appear.

    Your sub-agent response flow is working correctly; the second message is just the system showing internal reasoning by mistake.

    Current status

    There isn’t a workaround right now. Raise a support ticket so Microsoft can check whether your environment has moved to the updated orchestration runtime.

    Thank you.

    Karan Shewale.

    If this response resolves your issue, please Accept the answer and, if helpful, click the “Upvote” button. Your feedback helps us improve and assist others more effectively.

    0 comments

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.