trading-brain
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill functions as a coordination layer for a complex trading system. It establishes structured data contracts (MarketState, Signal, TradeResult) to ensure consistent information transfer between layers.
- [DATA_EXFILTRATION]: The skill references components for scraping social sentiment and news intelligence. While these involve network operations to fetch external data, the analysis found no evidence of exfiltrating sensitive local information, credentials, or environment variables to untrusted domains.
- [REMOTE_CODE_EXECUTION]: The implementation uses a local skill router for dynamic lookup of internal components. There are no instances of downloading and executing remote scripts or using dangerous functions like eval or exec on untrusted input.
- [PROMPT_INJECTION]: The skill architecture is susceptible to an Indirect Prompt Injection surface because it ingests untrusted data from news and social media sources (Layer 1). However, the system includes mitigation factors: it uses structured data for processing, implements a mandatory Risk Engine (Layer 5) with veto power, and requires multi-factor signal agreement (Layer 4) before action, which limits the impact of adversarial content in a single data source.
Audit Metadata