refactor

Pass

Audited by Gen Agent Trust Hub on Oct 29, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes user-provided source code files and directories to perform audits and refactoring tasks.
  • Ingestion points: Files and directories passed to audit-ui.ts and check-contrast.ts scripts, as well as files read using the view_file tool.
  • Boundary markers: None explicitly defined in the scripts for the content of the files being read.
  • Capability inventory: The skill uses view_file, replace_file_content, write_to_file, and run_command (to execute bun, npx, and bunx commands).
  • Sanitization: The audit scripts use regular expressions to detect patterns rather than executing the content of the files. The contrast checker parses specific hex color formats.
  • [COMMAND_EXECUTION]: The skill utilizes the run_command tool to execute auditing scripts and standard development tasks such as linting, type-checking, and testing using bun, npx, and bunx. These operations are consistent with the skill's primary purpose as a refactoring engine.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 29, 2026, 05:50 PM