new-module
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using
uv runfor code linting and testing in Step 8. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by interpolating user-provided input into shell commands.
- Ingestion points: The module name is provided via the
$ARGUMENTSvariable inSKILL.md. - Boundary markers: The skill lacks delimiters or instructions to treat the module name strictly as data, leaving it exposed to the execution context.
- Capability inventory: The skill has the capability to execute shell commands via the
uv runtool inSKILL.md. - Sanitization: The module name is used directly to construct file paths and shell commands in Step 8 without any validation, escaping, or sanitization, creating a risk of command injection if a malicious name is supplied.
Audit Metadata