images

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface due to its core functionality of reading and acting upon external file data.\n
  • Ingestion points: The skill reads image files (JPEG, PNG) and HTML files (src/**/*.html) from the filesystem.\n
  • Boundary markers: There are no defined delimiters or instructions to help the agent distinguish between data and potentially malicious instructions embedded in filenames or file content.\n
  • Capability inventory: The skill uses the Bash tool to execute optimization scripts, and the Write and Edit tools to manage file output.\n
  • Sanitization: No sanitization logic is provided to ensure that filenames or file contents are safe before being used as arguments in shell commands.\n- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute local Node.js scripts, such as node scripts/quality/optimize-images.js, using file paths as arguments. This could be exploitable if the environment does not properly handle or escape shell metacharacters in paths.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 06:11 AM