lint-new
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using
pnpm test-ciwith file paths constructed from the user-supplied$ARGUMENTSvariable. While this is a standard development workflow, using unvalidated user input in a shell context creates a potential command injection surface if the input contains shell metacharacters. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data (the rule name via
$ARGUMENTS) and incorporates it into file paths and shell commands. - Ingestion points: User-provided rule name ($ARGUMENTS) in
SKILL.md. - Boundary markers: None present; the variable is interpolated directly into paths and commands.
- Capability inventory: The skill performs file creation, file modification, and shell command execution (
pnpm). - Sanitization: There are no instructions provided to validate or sanitize the
$ARGUMENTSstring to prevent path traversal or command injection attacks.
Audit Metadata