FE-1725: Remember Petrinaut Voice disclosure for Live - #9748
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
1 Skipped Deployment
|
PR SummaryLow Risk Overview After the user acknowledges once, entering voice mode auto-starts Live (via extracted Regression tests add an in-memory Reviewed by Cursor Bugbot for commit c110233. Bugbot is set up for automated code reviews on this repo. Configure here. |
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 8b5995c. Configure here.
e8be5f2 to
8f313c3
Compare
lunelson
left a comment
There was a problem hiding this comment.
Requesting changes for the disclosure acknowledgement behavior:
-
Blocking: Live reuses
VOICE_INTERVIEW_DISCLOSURE_STORAGE_KEY, so a user who previously acknowledged the Realtime disclosure will auto-start Live without seeing Live-specific disclosure of the separate voice and transcription processing streams. Please use a Live-specific/version-bumped acknowledgement key, or first make the provider disclosures semantically equivalent before sharing one key. -
After an acknowledged Live auto-start fails,
LiveConversationControlrendersVoiceInterviewDisclosureagain withconsentedreset to false. This makes a returning user re-acknowledge the disclosure to retry. Please provide a distinct retry/error state or otherwise retain the acknowledged state instead of presenting consent again.
Optional naming suggestion: start returns whether startup was accepted, and that boolean controls retry after teardown. tryStartLiveConversation would make this contract and its provider scope clearer.
|
Fixed in |
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
761be96 to
c110233
Compare
lunelson
left a comment
There was a problem hiding this comment.
Re-reviewed the follow-up. Live now has a distinct acknowledgement key and a separate retry state, and the startup helper rename clarifies its boolean contract. The requested changes are resolved.

🌟 What is the purpose of this PR?
Make the Petrinaut Voice disclosure one-time for GPT-Live while keeping its consent separate from Realtime. Browser microphone permission remains separate.
🔗 Related links
🚫 Blocked by
🔍 What does this change?
Pre-Merge Checklist 🚀
🚢 Has this modified a publishable library?
This PR:
📜 Does this require a change to the docs?
The changes in this PR:
🕸️ Does this require a change to the Turbo Graph?
The changes in this PR:
None.
🐾 Next steps
None.
🛡 What tests cover this?
❓ How to test this?
📹 Demo
Not included; this behavior is covered by the regression tests.