specialist-engineering-expo-expert
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data by interpolating it into a sub-agent prompt.
- Ingestion points: User input is accepted through the
{the user's question}variable inSKILL.md. - Boundary markers: Absent; the user input is appended directly to the end of the system prompt for the spawned sub-agent without delimiters.
- Capability inventory: The sub-agent is designed for Expo framework guidance and uses the
Agenttool to operate. - Sanitization: There is no evidence of input sanitization or filtering before the data is passed to the sub-agent.
- [SAFE]: The skill instructions are consistent with its stated purpose as an Expo development assistant. No signs of data exfiltration, hardcoded credentials, obfuscation, or unauthorized remote code execution were found. All instructions follow legitimate software engineering best practices for the Expo ecosystem.
Audit Metadata