to-tasks
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Node.js script (
node .agents/scripts/cooking.mjs) with arguments provided by the user or derived from the context (<feature>). This pattern could be susceptible to command injection if the input is not properly sanitized. - [INDIRECT_PROMPT_INJECTION]: The skill processes content from multiple external documentation files, including
spec.md,goal.md, andACCEPTANCE.md. Malicious instructions embedded in these files could influence the agent's task generation or the arguments passed to local commands. - Ingestion points: Reads project-level documentation files like
spec.md,goal.md,ARCHITECTURE.md,CODE-MAP.md,DEV-STANDARDS.md,PROJECT.md, andACCEPTANCE.md(located in SKILL.md). - Boundary markers: None identified in the instructions to separate untrusted content from system instructions.
- Capability inventory: Can write new task files (
tasks/Pn.md) and execute a local Node.js script (cooking.mjs). - Sanitization: No explicit sanitization or validation of the content read from documentation files is mentioned.
Audit Metadata