yy-frontend-code-refine
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands like
git diffandgit stashto manage file discovery and ensure recovery paths are available. It also employsgrepandtestto check for specific dependencies within the local environment, and runsnpx prettierfor automated code formatting. - [EXTERNAL_DOWNLOADS]: The skill invokes
npx prettier, which may download the package from the official npm registry if it is not present locally. Prettier is a well-known service in the development community, and this behavior is standard for the tool's intended use. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted content from local source files. (1) Ingestion points: File content is scanned and processed in Step 1 and Step 3 of
SKILL.md. (2) Boundary markers: High-risk tasks in T03 and T04 require explicit user confirmation, and a 'no human in the loop' fallback skips these operations. (3) Capability inventory: The skill can read/write files and execute shell commands (git,npx). (4) Sanitization: The skill performs syntax validation usingprettier --checkafter modifications.
Audit Metadata