setup
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill incorporates user-provided data directly into its execution flow using the
{{ARGUMENTS}}placeholder without protective boundaries. - Ingestion points: User input enters the agent context through the
{{ARGUMENTS}}variable at the end ofSKILL.md. - Boundary markers: The skill lacks delimiters (such as XML tags or explicit markers) to isolate the user input from the system instructions.
- Capability inventory: The skill routes inputs to administrative tools including
/oh-my-claudecode:omc-setup,/oh-my-claudecode:omc-doctor, and/oh-my-claudecode:mcp-setup. - Sanitization: No validation or escaping is applied to the interpolated arguments to prevent potential instruction override or indirect injection attacks.
Audit Metadata