meme-engineering
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- PROMPT_INJECTION (SAFE): The skill uses natural instructional language to define a persona ('Meme Scientist'). While it contains instructions like 'Ignore generic approaches', these are context-specific constraints for domain expertise and do not attempt to bypass core AI safety filters or extract system prompts.
- DATA_EXFILTRATION (SAFE): No network operations, API calls, or sensitive file path references were detected. The skill operates entirely on text provided within the chat context.
- REMOTE_CODE_EXECUTION (SAFE): There are no scripts, package managers, or remote download commands (curl, wget) present in the analyzed files.
- OBFUSCATION (SAFE): All content is provided in clear-text markdown. No Base64, zero-width characters, or hidden unicode tags were identified.
- INDIRECT_PROMPT_INJECTION (LOW): The skill is designed to process user-provided meme content. While it lacks explicit boundary markers for untrusted data, the skill has no dangerous capabilities (like file writing or network access) that could be exploited via malicious user input.
Audit Metadata