meyers-effective-cpp
Fail
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: HIGHOBFUSCATIONPROMPT_INJECTION
Full Analysis
- [OBFUSCATION]: The skill uses zero-width Unicode characters to hide data within the Markdown headers of multiple files.
- In
SKILL.md, the H1 header# Scott Meyers Style Guidecontains 185 hidden characters consisting of Zero Width Space (U+200B), Zero Width Non-Joiner (U+200C), and Zero Width Joiner (U+200D). - In
references.md, the H1 header# Scott Meyers Referencescontains a similar block of 180 hidden characters. - The characters are organized into 8-bit sequences, which is a standard method for encoding binary data or hidden text within human-readable documents.
- [PROMPT_INJECTION]: The use of steganographic content in skill instructions is a high-risk indicator of prompt injection. This method allows for the delivery of hidden commands that can influence the agent's behavior, potentially overriding safety guidelines or security constraints without the user's knowledge. The deterministic static analysis correctly identified these sequences as steganographic prompt injection vectors.
Recommendations
- AI detected serious security threats
Audit Metadata