[BUG] Mobile: permission-prompt modal accepts accidental tap from unrelated screen interaction
Preflight Checklist
- [x] I have searched existing issues and this hasn't been reported yet
- [x] This is a single bug report (please file separate reports for different bugs)
- [x] I am using the latest version of Claude Code
What's Wrong?
The tool-permission modal (Allow/Deny) can render at the exact screen position of a tap
already in flight, before I've had a chance to read the prompt. The tap gets consumed as
an approval. This happened outside of typing/keyboard interaction, so it's a variant of
the keyboard-overlap case reported in #60059 and #67644 (both closed) — the underlying
problem (no guard against in-flight touches when the modal opens) is the same.
What Should Happen?
The modal's buttons should not be tappable for a short window (e.g. ~300-500ms) after it
renders, so a touch already in progress elsewhere on screen can never be misread as a
decision on the prompt.
Error Messages/Logs
N/A — no error message, just an unintended approval with no way to catch it after the fact.Steps to Reproduce
- Drive a Claude Code session from the Claude Android app.
- Tap/interact anywhere on screen right as the agent triggers a tool call needing approval.
- The permission modal opens under the in-flight tap and immediately registers "Approve".
Claude Model
Not sure / Multiple models
Is this a regression?
I don't know
Last Working Version
_No response_
Claude Code Version
Android app - Claude 1.260904.19
Platform
Anthropic API
Operating System
Windows
Terminal/Shell
IntelliJ IDEA terminal
Additional Information
There is a duplicate for iPhone app, which is closed due to inactivity:
- #81250