tzai-wechat
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a shell script resolved from several potential installation paths within the user's home directory (e.g.,
~/.agents/skills/tzai-image/scripts/tzai-image). - [EXTERNAL_DOWNLOADS]: If the required engine is missing, the skill provides a command to download and install a vendor-owned dependency:
npx skills add kedoupi/tzai-image-skill -g --skill tzai-image -y. - [PROMPT_INJECTION]: User-provided text is interpolated directly into a bash command line via the
--prompt "<user subject>"argument. This presents an indirect injection surface where malicious user input could attempt to escape the command string and execute arbitrary shell commands if the agent does not perform adequate sanitization before execution.
Audit Metadata