tzai-food

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill implements a search routine to locate an executable script named 'tzai-image' across several standard agent-skill directories in the user's home folder (e.g., ~/.agents/, ~/.claude/, ~/.cursor/). If found, it executes this script using bash.
  • [COMMAND_EXECUTION]: The skill interpolates user-provided text directly into a shell command: bash "$ENGINE" food --prompt "<user subject>". This pattern presents a potential command injection surface if the user-supplied '' contains shell metacharacters that are not correctly escaped or handled by the underlying engine script.
  • [EXTERNAL_DOWNLOADS]: If the required engine is not found, the skill prompts the user to download and install the dependency using npx skills add kedoupi/tzai-image-skill. This dependency originates from the same author ('kedoupi') and is a standard part of the tool's ecosystem.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 02:50 AM
Security Audit — agent-trust-hub — tzai-food