Add orchestration message display setting#12219
Conversation
Co-Authored-By: Oz <oz-agent@warp.dev>
4650599 to
51a9569
Compare
|
I'm starting a first review of this pull request. You can view the conversation on Warp. I completed the review and no human review was requested for this pull request. Comment Powered by Oz |
There was a problem hiding this comment.
Overview
This PR adds a new AI setting for orchestration message expansion behavior, wires it into settings/command-palette state, updates collapsible state handling for sent and received orchestration messages, and switches orchestration message bodies to Markdown-backed rendering.
Concerns
- Markdown-rendered orchestration message bodies now inherit the standard executable shell-code controls, so child-agent transcript content can surface runnable commands where it was previously inert text.
Security
- A child-agent or orchestrator message containing a fenced shell snippet can now expose the normal run action through the transcript renderer. That should be disabled or explicitly opted into for orchestration message bodies.
Verdict
Found: 0 critical, 1 important, 0 suggestions
Request changes
Comment /oz-review on this pull request to retrigger a review (up to 3 times on the same pull request).
Powered by Oz
c81a895 to
e9ec603
Compare
e9ec603 to
44d3696
Compare

Description
Right now orchestration messages auto-collapse so users have to manually click to expand them. It makes sense to keep this as the default option (for now), but I think there should be a setting for the message expansion behavior, similar to thinking traces.
The setting has 3 options:
Testing
https://www.loom.com/share/884b5253a20f4be3bb9fc1db40834b46
Conversation: https://staging.warp.dev/conversation/350fa0b0-68bd-4f95-9c58-cfdfae17d0c5
Run: https://oz.staging.warp.dev/runs/019e9444-002f-7b9c-91ec-9ff5eb8c8420
This PR was generated with Oz.