nichols-practical-rust
Fail
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: HIGHPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill's primary header contains a sequence of hidden Unicode characters (Zero-Width Space
\u200B, Zero-Width Non-Joiner\u200C, and Zero-Width Joiner\u200D) positioned after the visible text. This technique, known as steganography, allows information to be embedded in text without being visible to the user. Decoding the binary pattern (where\u200Brepresents 0 and\u200Crepresents 1) reveals a structured payload starting with 'SK1L' followed by additional binary data. This type of obfuscation is used to bypass security filters, hide hidden instructions, or track skill usage in an unauthorized manner.
Recommendations
- AI detected serious security threats
Audit Metadata