beforeSubmitPrompt hook block is broken

Support Bug Reports
,
1

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

When beforeSubmitPrompt hook blocked the message it is not sent to the model at first but it still remains in the chat UI. If I edit it and resend, that makes sense. But if I leave it unchanged and send a new message instead, the previously blocked message gets sent together with the new one.

Steps to Reproduce

  1. Send a prompt that gets blocked by beforeSubmitPrompt hook.
  2. Leave the blocked message unchanged in the chat.
  3. Send a new message in the same conversation.
  4. Notice that the previously blocked message is sent together with the new message.

Expected Behavior

A blocked message should not be sent to the model unless the user explicitly edits and resubmits it. Sending a new message should not cause a previously blocked message to be sent.

Screenshots / Screen Recordings

image (2).png
image (2).png920×1356 90.3 KB

Operating System

MacOS

Version Information

Version: 3.0.12 (Universal)
VSCode Version: 1.105.1
Commit: a80ff7dfcaa45d7750f6e30be457261379c29b00
Date: 2026-04-04T00:13:18.452Z
Layout: editor
Build Type: Stable
Release Track: Default
Electron: 39.8.1
Chromium: 142.0.7444.265
Node.js: 22.22.1
V8: 14.2.231.22-electron.0
OS: Darwin arm64 25.4.0

Does this stop you from using Cursor

Yes - Cursor is unusable

4

Hey, thanks for the report. I can see the screenshot showing PII detection, and I get that this is a serious issue for a security-critical use case.

This is a known bug. When beforeSubmitPrompt returns continue: false, the blocked message still stays in the conversation history and gets sent on the next turn. The team is aware, and your report helps with prioritization.

Current workaround: after a block, start a new chat. This clears the context and prevents the blocked content from leaking. Not ideal, but it’s the only reliable option right now.

Related thread for context: Blocked messages (BeforeSubmitPrompt hook returns continue: false) are still included in later LLM context/history

Let me know if you have any questions.