claude-code - 💡(How to fix) Fix Permission prompt for Skill tool should show which skill is being invoked [1 comments, 2 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#51883Fetched 2026-04-23 07:42:21
View on GitHub
Comments
1
Participants
2
Timeline
5
Reactions
0
Author
Timeline (top)
labeled ×2closed ×1commented ×1unsubscribed ×1

When Claude invokes the Skill tool, the permission prompt displays:

Allow Claude to use Skill ?

This gives the user no context about which skill is about to run. Compare this to the Bash tool, which shows the actual command:

Allow Claude to run lsof -ti:3000 ?

Root Cause

When Claude invokes the Skill tool, the permission prompt displays:

Allow Claude to use Skill ?

This gives the user no context about which skill is about to run. Compare this to the Bash tool, which shows the actual command:

Allow Claude to run lsof -ti:3000 ?

RAW_BUFFERClick to expand / collapse

Summary

When Claude invokes the Skill tool, the permission prompt displays:

Allow Claude to use Skill ?

This gives the user no context about which skill is about to run. Compare this to the Bash tool, which shows the actual command:

Allow Claude to run lsof -ti:3000 ?

Expected behavior

The prompt should include the skill name, e.g.:

Allow Claude to use Skill(editor-start) ?

This is consistent with how Bash already surfaces its argument, and gives users the context they need to make an informed allow/deny decision — especially when skills are invoked automatically (e.g. via a CLAUDE.md instruction at session start).

Why it matters

Skills can be invoked automatically without any user action (e.g. a global ~/.claude/CLAUDE.md that says "run /editor-start at the start of every conversation"). When that happens, the user sees a generic "Skill ?" prompt with no indication of what's about to run. The only safe response is to allow everything or investigate manually — neither is great UX.

Suggested fix

Include the skill/command name in the permission prompt display string, the same way Bash includes its command argument. The tool already receives the skill name as a parameter, so this should be a small display-layer change.

extent analysis

TL;DR

Modify the permission prompt display string to include the skill name, similar to how the Bash tool includes its command argument.

Guidance

  • Review the code responsible for generating the permission prompt for the Skill tool to identify where the display string is constructed.
  • Update the display string to include the skill name, which is already received as a parameter, to provide users with the necessary context.
  • Verify that the updated prompt correctly displays the skill name for both manually and automatically invoked skills.
  • Consider reviewing the Bash tool's implementation for inspiration on how to incorporate the command argument into the prompt.

Example

No code snippet is provided as the issue does not include specific code references, but the fix should involve modifying the string that generates the permission prompt to include the skill name, e.g., Allow Claude to use **Skill({skillName})** ?.

Notes

The suggested fix assumes that the skill name is already available as a parameter when generating the permission prompt. If this is not the case, additional changes may be necessary to pass the skill name to the prompt generation code.

Recommendation

Apply the workaround by modifying the permission prompt display string to include the skill name, as this provides a clear and consistent user experience similar to the Bash tool.

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

The prompt should include the skill name, e.g.:

Allow Claude to use Skill(editor-start) ?

This is consistent with how Bash already surfaces its argument, and gives users the context they need to make an informed allow/deny decision — especially when skills are invoked automatically (e.g. via a CLAUDE.md instruction at session start).

Still need to ship something?

×6

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

Back to top recommendations

TRENDING