specialist-engineering-frontend-developer
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by interpolating raw user input directly into the subagent's system prompt without the use of boundary markers or isolation instructions.
- Ingestion points: User input is interpolated via the
{the user's question}placeholder at the end of the subagent prompt inSKILL.md. - Boundary markers: None present; the user's question is appended directly to the instructions.
- Capability inventory: The subagent is designed to generate and optimize frontend code (React, Vue, Angular) and propose architectural patterns.
- Sanitization: No evidence of input sanitization or filtering of instructions within the interpolated data.
Audit Metadata