Repository navigation
Invoke the skill from a $code-review-ad prompt mention - #10
Merged
Merged
Conversation
In Claude Code a plain-text mention of the skill leaves invocation to the model, and Codex-style $code-review-ad mentions mean nothing. Ship a UserPromptSubmit hook with the Claude plugin. When a prompt mentions $code-review-ad or $code-review-ad:code-review-ad, it adds context telling Claude to invoke the skill with the Skill tool first. Synchronize the plugin version at 1.1.0. Fixes #9
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fixes #9
Issue
In Claude Code, only a leading
/code-review-ad:code-review-adslash command reliably runs the skill. A plain-text mention leaves invocation to the model, and Codex-style$code-review-admentions mean nothing.Fix
The Claude plugin now registers a
UserPromptSubmithook (claude/hooks/hooks.json,claude/hooks/skill_mention.py). When a prompt mentions$code-review-ador$code-review-ad:code-review-adanywhere, the hook adds context telling Claude to invoke the skill with the Skill tool before doing anything else. Other prompts and malformed hook input pass through with no output.Version synchronized at 1.1.0; README and PROVENANCE updated.
Testing
python3 scripts/check.py,python3 scripts/test.py(92 passed, including newtests/test_skill_mention.py)claude plugin validate .passes (only the existing marketplace-description and icon warnings)claude -p --plugin-dir .: a prompt with$code-review-adreceived the hook context; a prompt with a plaincode-review-admention did not