langchain-ai/deepagentsjs
1
0
Fork 0
mirror of https://github.com/langchain-ai/deepagentsjs.git synced 2026-07-18 16:04:44 -04:00
Code Issues 42 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #40] Cannot Prevent Subagents to show output to chat interface #12

New Issue
Closed
opened 2026-02-16 06:16:48 -05:00 by yindo · 4 comments
yindo commented 2026-02-16 06:16:48 -05:00
Owner
Copy Link
Copy Source

Originally created by @amitchaudharyc on GitHub (Oct 3, 2025).
Original GitHub issue: https://github.com/langchain-ai/deepagentsjs/issues/40

I am building chat interface and the issue is when subagents work, they produces their own output, which generally I want it to be passed to main agent without showing it to the user. I can filter it in the frontend but that's not what I want. I want to disable it in graph itself.

Originally created by @amitchaudharyc on GitHub (Oct 3, 2025). Original GitHub issue: https://github.com/langchain-ai/deepagentsjs/issues/40 I am building chat interface and the issue is when subagents work, they produces their own output, which generally I want it to be passed to main agent without showing it to the user. I can filter it in the frontend but that's not what I want. I want to disable it in graph itself.
yindo commented 2026-02-16 06:16:48 -05:00
Author
Owner
Copy Link
Copy Source

@christian-bromann commented on GitHub (Jan 3, 2026):

@amitchaudharyc thanks for flagging this, I will look into rendering techniques for deepagents. We are actively working on improvements in this area.

@christian-bromann commented on GitHub (Jan 3, 2026): @amitchaudharyc thanks for flagging this, I will look into rendering techniques for deepagents. We are actively working on improvements in this area.
yindo commented 2026-02-16 06:16:48 -05:00
Author
Owner
Copy Link
Copy Source

@christian-bromann commented on GitHub (Feb 5, 2026):

fyi, you can follow progress here: https://github.com/langchain-ai/langgraphjs/pull/1903

@christian-bromann commented on GitHub (Feb 5, 2026): fyi, you can follow progress here: https://github.com/langchain-ai/langgraphjs/pull/1903
yindo commented 2026-02-16 06:16:48 -05:00
Author
Owner
Copy Link
Copy Source

@chopraj commented on GitHub (Feb 5, 2026):

Not sure how helpful this is but I used tags on the subagents and filtered out the events that came from these agents on the backend.

@chopraj commented on GitHub (Feb 5, 2026): Not sure how helpful this is but I used tags on the subagents and filtered out the events that came from these agents on the backend.
yindo changed title from Cannot Prevent Subagents to show output to chat interface to [GH-ISSUE #40] Cannot Prevent Subagents to show output to chat interface 2026-06-05 17:20:56 -04:00
yindo closed this issue 2026-06-05 17:20:57 -04:00
yindo commented 2026-06-05 17:20:57 -04:00
Author
Owner
Copy Link
Copy Source

@christian-bromann commented on GitHub (May 18, 2026):

Thanks for raising this. The underlying need makes sense: subagent output is useful for debugging and coordination, but it should not automatically appear as end-user assistant text.

The new streaming primitives address this by separating channel and namespace concerns. The main chat can subscribe to root-agent messages, while subagent messages remain available under subagent namespaces or a dedicated subagent projection. That lets the app decide whether to hide subagent text entirely, render it in a debug panel, or summarize it as progress.

The important distinction is that Deep Agents does not need to suppress all subagent output at the graph level. In the new v3 stream surface, subagent activity is exposed with enough identity for consumers to route it correctly.

The relevant cookbook examples are here: streaming subagents examples

With agent.streamEvents(..., { version: "v3" }), consumers can use run.subagents to observe subagent messages, tool calls, task input, and output separately from the root agent.

<!-- gh-comment-id:4477612598 --> @christian-bromann commented on GitHub (May 18, 2026): Thanks for raising this. The underlying need makes sense: subagent output is useful for debugging and coordination, but it should not automatically appear as end-user assistant text. The new streaming primitives address this by separating channel and namespace concerns. The main chat can subscribe to root-agent `messages`, while subagent messages remain available under subagent namespaces or a dedicated subagent projection. That lets the app decide whether to hide subagent text entirely, render it in a debug panel, or summarize it as progress. The important distinction is that Deep Agents does not need to suppress all subagent output at the graph level. In the new v3 stream surface, subagent activity is exposed with enough identity for consumers to route it correctly. The relevant cookbook examples are here: [streaming subagents examples](https://github.com/langchain-ai/streaming-cookbook/tree/main/typescript/streaming/src/subagents) With `agent.streamEvents(..., { version: "v3" })`, consumers can use `run.subagents` to observe subagent messages, tool calls, task input, and output separately from the root agent.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
enhancement
help wanted
needs-response
pull-request
Mirrored from GitHub Pull Request
 stale
triage: high-impact
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/deepagentsjs#12
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?