claude-code - 💡(How to fix) Fix Approved permission prompts not surfaced to Claude in conversation [1 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
anthropics/claude-code#45490Fetched 2026-04-09 08:04:12
View on GitHub
Comments
0
Participants
1
Timeline
4
Reactions
0
Participants
Timeline (top)
labeled ×4

When a user approves a permission prompt in Claude Code, Claude has no visibility into what was approved. Claude only sees tool call success/failure — not which permission was prompted or that an approval occurred.

Error Message

  • Claude sees only the tool result (success or error)

Root Cause

  • User sees a permission prompt and approves it
  • Claude sees only the tool result (success or error)
  • Claude has no awareness that a prompt occurred or what triggered it
  • Claude cannot distinguish "this worked because it was already allowed" from "this worked because user just approved it"
RAW_BUFFERClick to expand / collapse

Summary

When a user approves a permission prompt in Claude Code, Claude has no visibility into what was approved. Claude only sees tool call success/failure — not which permission was prompted or that an approval occurred.

Expected behavior

When a user approves a permission prompt, Claude should receive some signal — either the prompt text, the tool + arguments that triggered it, or a notification that approval happened. This would allow Claude to:

  1. Proactively add the permission to settings so it doesn't prompt again
  2. Accurately diagnose permission issues rather than guessing

Current behavior

  • User sees a permission prompt and approves it
  • Claude sees only the tool result (success or error)
  • Claude has no awareness that a prompt occurred or what triggered it
  • Claude cannot distinguish "this worked because it was already allowed" from "this worked because user just approved it"

Impact

  • Users have to manually describe what prompted them
  • Claude guesses at root cause instead of diagnosing accurately
  • Approved permissions aren't automatically reflected in settings.json

Environment

  • Claude Code CLI (VSCode extension)
  • macOS

🤖 Filed via Claude Code

extent analysis

TL;DR

Modify the Claude Code CLI to capture and send permission prompt approvals to Claude, allowing it to update settings and diagnose issues accurately.

Guidance

  • Investigate the permission prompt mechanism in the Claude Code CLI to determine how to capture approval events and send relevant data to Claude.
  • Consider adding a notification or callback mechanism to inform Claude when a permission prompt is approved, including details about the prompt or triggering tool and arguments.
  • Review the current implementation of tool call success/failure handling in Claude to identify where the approval data can be integrated and utilized.
  • Explore updating the settings.json file automatically when a permission is approved, to reflect the new permission state.

Example

No specific code example can be provided without more details on the Claude Code CLI implementation, but a potential approach might involve adding an event listener or callback function to capture permission prompt approvals and send a notification to Claude.

Notes

The solution may require modifications to the Claude Code CLI and/or the Claude backend to handle the new approval data, and careful consideration should be given to ensuring that the approval data is handled correctly and securely.

Recommendation

Apply a workaround by modifying the Claude Code CLI to capture and send permission prompt approvals to Claude, as upgrading to a fixed version is not mentioned as an option in the issue. This will allow Claude to update settings and diagnose issues accurately, improving the overall user experience.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

FAQ

Expected behavior

When a user approves a permission prompt, Claude should receive some signal — either the prompt text, the tool + arguments that triggered it, or a notification that approval happened. This would allow Claude to:

  1. Proactively add the permission to settings so it doesn't prompt again
  2. Accurately diagnose permission issues rather than guessing

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING