new-user
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interpolates user-provided arguments directly into a shell command, creating an injection surface.
- Ingestion points: User-provided
<email>and<password>arguments parsed from the instruction inSKILL.md. - Boundary markers: Absent. There are no clear delimiters or validation rules specified to separate input data from the command context.
- Capability inventory: Executes arbitrary text interpolation into a shell command via
npx medusa user -e <email> -p <password>using the Bash tool. - Sanitization: Absent. The instructions do not mandate validation or sanitization of the input arguments before execution.
Audit Metadata