explore
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Node.js script using the command
node .agents/scripts/cooking.mjs status <feature>. The<feature>parameter, which is provided by the user, is interpolated directly into the command, creating a potential command injection surface if the input is not sanitized by the underlying platform.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user requirements and task identifiers to generate and managegoal.mdfiles.\n - Ingestion points: User-supplied requirement descriptions and the
<feature>unit name (SKILL.md).\n - Boundary markers: The instructions do not define delimiters or validation steps for external input before processing.\n
- Capability inventory: The skill possesses file system access (read, write, and delete operations in
.agents/cooking/) and subprocess execution capabilities (node).\n - Sanitization: There are no instructions to escape or sanitize user-provided content before it is interpolated into files or shell commands.
Audit Metadata