quality-posters
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on executing a local script
generate.jsthrough shell commands to perform its primary function. - [PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection or command injection. The agent is instructed to pass a user-provided brief directly into a shell command template:
node generate.js "<brief>". This pattern is vulnerable if the input contains shell metacharacters like backticks or semicolons. - Ingestion points: The
<brief>argument provided by the user (SKILL.md). - Boundary markers: The brief is wrapped in double quotes in the documentation, but this is an insufficient security boundary for shell execution.
- Capability inventory: Execution of local scripts via
node(SKILL.md). - Sanitization: The instructions lack any requirement for the agent to sanitize or validate the user-supplied brief before execution.
Audit Metadata