bulk-remove-classname

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (MEDIUM): The skill utilizes npx classname-search, which downloads and executes code from the npm registry at runtime. The package classname-search does not originate from a trusted organization defined in the security policy (e.g., Anthropic, Google, Microsoft, Vercel), making it an unverifiable dependency.
  • [COMMAND_EXECUTION] (LOW): The skill executes multiple shell commands (npx, rm) using user-provided inputs (<glob>, <class-name>). While the instructions use single quotes to mitigate simple command injection, the execution of external tools on local files remains a risk if the tool itself has vulnerabilities.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill processes contents of HTML/JSX files which could contain malicious instructions.
  • Ingestion points: Files matching the user-provided <glob> pattern are read by the classname-search tool.
  • Boundary markers: Absent; there are no specific markers or instructions to ignore LLM-targeted content within the target files.
  • Capability inventory: The skill can read files, modify files (remove class names), and delete files (rm 'verification-<class-name>.jsonl').
  • Sanitization: Absent; the skill does not specify any sanitization or validation of the content found within the target files before processing.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:21 PM