specialist-engineering-code-reviewer
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it interpolates raw user input into a sub-agent's system prompt without delimiters or sanitization.
- Ingestion points: The user's input variable
{the user's question}is directly embedded in the sub-agent prompt withinSKILL.md. - Boundary markers: There are no markers or instructions to isolate user input or prevent it from overriding the agent's core instructions.
- Capability inventory: The skill uses the
Agenttool to spawn sub-agents but does not specify file-write or network capabilities in the provided instructions. - Sanitization: No input validation, escaping, or filtering is applied to the user-provided question.
Audit Metadata