product-engineer-agent
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- Command Execution (LOW): The industrial-designer.md and mechanical-engineer.md files provide templates for the agent to generate and execute shell commands (python3) to call an image generation script. This pattern is common for inter-skill communication but can be vulnerable to command injection if user-supplied inputs like product names contain shell metacharacters.
- Indirect Prompt Injection (LOW): The skill ingests untrusted product descriptions and interpolates them into prompt templates for downstream tools. Evidence Chain: 1. Ingestion points: [PRODUCT] and [ENVIRONMENT DESCRIPTION] fields in industrial-designer.md and mechanical-engineer.md. 2. Boundary markers: Absent; inputs are directly placed into command strings. 3. Capability inventory: Execution of Python scripts through shell commands. 4. Sanitization: No sanitization or escaping instructions are provided in the agent definitions.
Audit Metadata