tzai-xhs

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to install a required engine using the command npx skills add kedoupi/tzai-image-skill. This involves downloading and executing a package associated with the skill's author.
  • [COMMAND_EXECUTION]: The skill searches for a shell script named tzai-image within several hidden application directories in the user's home folder. If found, it executes this script with parameters including user-provided prompts and local file paths.
  • [PROMPT_INJECTION]: The skill processes user-provided subject matter to generate visual content, representing an indirect prompt injection surface.
  • Ingestion points: The user's subject/theme input provided via the --prompt argument in SKILL.md.
  • Boundary markers: None explicitly defined in the execution logic; however, the skill references a compile-guide.md for structuring inputs.
  • Capability inventory: Execution of local shell scripts and file writing (.png assets) to the local filesystem.
  • Sanitization: No explicit sanitization of user-provided prompt strings is present in the provided shell logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 02:20 AM
Security Audit — agent-trust-hub — tzai-xhs