ljg-word-flow
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill orchestrates the execution of
ljg-wordandljg-cardtools. These are identified as legitimate internal resources belonging to the author 'lijigang'.\n- [DATA_EXFILTRATION]: Instructions specify saving generated PNG files to the local~/Downloads/directory. This file system access is standard for a tool designed to produce visual output for the user.\n- [PROMPT_INJECTION]: The skill ingests user-provided text to be used as input for word analysis. This follows a standard data ingestion pattern for word-processing utilities.\n - Ingestion points: User-provided English words in the command arguments.\n
- Boundary markers: Not explicitly defined in the skill instructions.\n
- Capability inventory: Execution of
ljg-wordandljg-cardtools; file write access to~/Downloads/.\n - Sanitization: Not explicitly mentioned in the skill body.
Audit Metadata