scarface-mean-reversion
Fail
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: HIGHPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The H1 header in
SKILL.mdcontains a sequence of zero-width characters (ZWSP, ZWNJ, ZWJ) following the visible text. This obfuscated payload encodes a character sequence (starting with 'SK1LL') that is hidden from the user but can be interpreted by the AI model. This is a common vector for stealthy prompt injection intended to bypass safety filters or provide hidden instructions.\n- [SAFE]: The Python code snippets provided for mean reversion scanning and position management use standard libraries likepandas,numpy, andscipy. The code performs mathematical calculations for trading strategies and does not contain malicious patterns such as network exfiltration, unauthorized file system access, or privilege escalation.
Recommendations
- AI detected serious security threats
Audit Metadata