llm-trading-agent-security
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a security guide, providing regex-based patterns to detect and prevent prompt injection attacks in a financial context.
- [SAFE]: Secret management practices align with industry standards by recommending the use of environment variables for private keys rather than hardcoding.
- [SAFE]: External network references are limited to well-known and legitimate blockchain infrastructure (Flashbots RPC) used for transaction protection.
- [SAFE]: The logic for spend limits and circuit breakers provides essential safety rails for autonomous agents without introducing malicious side effects.
Audit Metadata