The Agent Skills Directory

[PROMPT_INJECTION]: Indirect prompt injection vulnerability surface detected. The skill processes external, potentially untrusted code files from the src directory which could contain malicious instructions designed to influence the review outcome. * Ingestion points: Files with extensions .vue, .js, .css, .scss, and .less within the src/ directory. * Boundary markers: Analysis is guided by specific dimension IDs (D01-D09) and external reference documents in the references/ folder. * Capability inventory: The skill uses git diff to identify files and has the capability to automatically trigger a subsequent skill, yy-frontend-commit, upon a successful review. * Sanitization: No explicit sanitization or filtering of code comments or content is performed prior to the analysis, though instructions state that comment-related issues are ignored.
[COMMAND_EXECUTION]: The skill uses git diff to automatically identify changed files in the repository. This command execution is used solely for determining the scope of the code review.

yy-frontend-vue2-review