pike-simplicity-first
Fail
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: HIGHPROMPT_INJECTIONOBFUSCATION
Full Analysis
- [OBFUSCATION]: Large blocks of zero-width Unicode characters, including Zero Width Space (U+200B), Zero Width Non-Joiner (U+200C), Zero Width Joiner (U+200D), and Word Joiner (U+2060), are embedded within the primary headers of both
SKILL.mdandreferences/philosophy.md. These characters are invisible in standard markdown rendering but are processed by the AI agent during skill ingestion. - [PROMPT_INJECTION]: The hidden Unicode character sequences in the document headers encode steganographic instructions designed to bypass user oversight and manipulate the AI agent's internal state or behavior. The use of steganography within instructional content is a high-confidence indicator of malicious intent to perform prompt injection.
Recommendations
- AI detected serious security threats
Audit Metadata