netflix-chaos-engineering
Fail
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: HIGHOBFUSCATIONPROMPT_INJECTION
Full Analysis
- [OBFUSCATION]: The headers in
SKILL.mdandreferences/principles.mdcontain extensive sequences of zero-width characters (U+200B, U+200C, U+200D). This is a form of Unicode steganography designed to hide content from human review while ensuring it is processed by the AI agent.\n - Evidence in
SKILL.md: A hidden sequence of zero-width characters follows the '# Netflix Chaos Engineering' header.\n - Evidence in
references/principles.md: A hidden sequence of zero-width characters follows the '# Principles of Chaos Engineering' header.\n- [PROMPT_INJECTION]: The hidden Unicode characters in the document headers constitute a steganographic prompt injection attack. These hidden payloads are typically used to deliver instructions that bypass agent constraints, override system instructions, or exfiltrate data without the user's knowledge. The use of zero-width characters in primary headers is a high-confidence indicator of malicious intent to subvert agent behavior.
Recommendations
- AI detected serious security threats
Audit Metadata