skill-ab

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local JavaScript file (ab-test.mjs) via Node.js to manage testing comparisons.\n- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection by processing external file paths and outputting results without summarization or modification.\n
  • Ingestion points: The version-b-path argument in the command interface.\n
  • Boundary markers: No delimiters or instructions to disregard embedded commands are included.\n
  • Capability inventory: The skill can execute code via Node.js and access the local filesystem through provided arguments.\n
  • Sanitization: No sanitization or path validation is specified within the skill metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 08:17 AM