Repository navigation
Conversation
Skill token patterns required an ASCII letter first even though discovery accepts digit-leading directory names. Allow digits at the first position in composer parsing, web and mobile rendering, and Claude dispatch.
github-actions
Bot
added
vouch:unvouched
![macroscopeapp[bot]](https://avatars.githubusercontent.com/in/900172?s=60&v=4){kind=small}
| * dispatched skill are always the same set. | ||
| */ | ||
| const SKILL_MENTION_PATTERN = /(^|\s)\$([a-zA-Z][a-zA-Z0-9:_-]*)(?=\s|$)/g; | ||
| const SKILL_MENTION_PATTERN = /(^|\s)\$([a-zA-Z0-9][a-zA-Z0-9:_-]*)(?=\s|$)/g; |
There was a problem hiding this comment.
🟡 Medium Drivers/ClaudeSkillDispatch.ts:32
Use $1 now is dispatched as skill 1 and rewritten to /1 now, even though the composer treats $1 as plain text. The pattern permits a digit as the first character; require an alphabetic first character to keep numeric expressions from triggering skills.
| const SKILL_MENTION_PATTERN = /(^|\s)\$([a-zA-Z0-9][a-zA-Z0-9:_-]*)(?=\s|$)/g; | |
| const SKILL_MENTION_PATTERN = /(^|\s)\$([a-zA-Z][a-zA-Z0-9:_-]*)(?=\s|$)/g; |
🤖 Copy this AI Prompt to have your agent fix this:
In file @apps/server/src/provider/Drivers/ClaudeSkillDispatch.ts around line 32:
`Use $1 now` is dispatched as skill `1` and rewritten to `/1 now`, even though the composer treats `$1` as plain text. The pattern permits a digit as the first character; require an alphabetic first character to keep numeric expressions from triggering skills.
|
Caution Review failedThe pull request is closed. ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Advanced Run ID: 📒 Files selected for processing (7)
📝 WalkthroughWalkthroughSkill token and mention matching now accepts names that start with digits. Shared tokenization excludes purely numeric dollar expressions. Web, mobile, and Claude dispatch tests cover the updated behavior. ChangesNumeric skill-name support
Estimated code review effort: 2 (Simple) | ~10 minutes Change: Bug fix Sequence Diagram(s)sequenceDiagram
participant Composer
participant SkillTokenizer
participant ClientRenderer
participant ClaudeDispatcher
Composer->>SkillTokenizer: Submit "$2spec review this"
SkillTokenizer-->>ClientRenderer: Return skill token "2spec"
Composer->>ClaudeDispatcher: Submit "$2spec review this"
ClaudeDispatcher-->>Composer: Return "/2spec review this"
Suggested reviewers: ✨ Finishing Touches 💡 1🛠️ Fix failing CI checks 💡
🧪 Generate unit tests (beta)
Comment |
ApprovabilityVerdict: Would Approve Macroscope's review found this PR approvable — This is a small, tested parsing fix that extends skill recognition to names beginning with digits across the composer, clients, and Claude dispatch path, without schema, infrastructure, or default changes. A remaining Medium correctness finding concerns numeric-only Not approved because:
Adjust the Minimum Blocking Severity for this repo — including turning it Off — in Settings. You can add or adjust custom eligibility rules. Learn more. |
1 participant
Closed: this PR accidentally selected an unrelated existing fork branch after a branch-name collision. The verified fix for #12061 is in #12078.
Summary by CodeRabbit
$2spec, are now recognized and processed consistently across chat, markdown, and command dispatch.$1and$100continue to be treated as plain text rather than skill references.