Issue Station
Sign In
Tools hereToolbox

claude-code - 💡(How to fix) Fix [BUG] Skills with `disable-model-invocation: true` intermittently fail on user invocation in VS Code extension

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…

Error Message

Error Messages/Logs

Fix Action

Fix / Workaround

Workaround: Remove disable-model-invocation: true — works but the model may auto-trigger it.

Code Example



---

---
   name: my-skill
   description: "A skill"
   user-invocable: true
   disable-model-invocation: true
   ---
RAW_BUFFERClick to expand / collapse

Preflight Checklist

  • I have searched existing issues and this hasn't been reported yet
  • This is a single bug report (please file separate reports for different bugs)
  • I am using the latest version of Claude Code

What's Wrong?

In the VS Code extension, skills with disable-model-invocation: true and user-invocable: true intermittently fail when explicitly invoked by the user via /command.

Observed behavior: Intermittently, skills with disable-model-invocation: true are not available for user invocation via /command. The skill either doesn't appear in the skills list provided to the model, or silently fails to invoke.

The issue does not always reproduce — sometimes the skill works correctly, sometimes it fails in the next session.

Expected behavior: disable-model-invocation should only prevent the model from auto-triggering the skill. Explicit user invocation via /command should always work regardless of this flag.

Workaround: Remove disable-model-invocation: true — works but the model may auto-trigger it.

Claude Model

Opus

Claude Code for VS Code version

2.1.107

Operating System

Ubuntu/Debian Linux

Additional Information

Previously reported as #24042 (auto-closed as stale, locked) and #21649 (closed as duplicate). The underlying issue remains unresolved.

What Should Happen?

disable-model-invocation should only prevent the model from auto-triggering the skill. Explicit user invocation via /command should always work regardless of this flag.

Error Messages/Logs

Steps to Reproduce

  1. Create a skill with frontmatter: 
    ---
name: my-skill
description: "A skill"
user-invocable: true
disable-model-invocation: true
---
  2. Type /my-skill in VS Code Claude Code extension
  3. Skill appears in autocomplete but fails on invocation

Claude Model

Opus

Is this a regression?

I don't know

Last Working Version

No response

Claude Code Version

2.1.107

Platform

Anthropic API

Operating System

Ubuntu/Debian Linux

Terminal/Shell

Other

Additional Information

No response

extent analysis

TL;DR

The issue can be worked around by removing disable-model-invocation: true from the skill configuration, but a more targeted fix is needed to ensure the model does not auto-trigger the skill while still allowing explicit user invocation.

Guidance

  • Verify that the issue is indeed related to the disable-model-invocation flag by testing the skill with this flag removed.
  • Investigate the logic behind the disable-model-invocation flag to understand why it's preventing explicit user invocation via /command.
  • Consider adding a separate flag or configuration option to distinguish between auto-triggering and explicit user invocation.
  • Test the skill with different invocation methods to see if the issue is specific to the /command method.

Example

---
name: my-skill
description: "A skill"
user-invocable: true
auto-trigger: false
---

This example shows a potential configuration change to add a separate auto-trigger flag, which could allow for more fine-grained control over skill invocation.

Notes

The issue may be related to the interaction between the disable-model-invocation flag and the user-invocable flag. Further investigation is needed to determine the root cause of the issue.

Recommendation

Apply workaround: Remove disable-model-invocation: true from the skill configuration, as this has been shown to resolve the issue, although it may not be the ideal solution. A more targeted fix is needed to ensure the model does not auto-trigger the skill while still allowing explicit user invocation.

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…
#api#tensor shape#autograd error#model save/load#optimization

Still need to ship something?

×6

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

Back to top recommendations

TRENDING