joke-engineering
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze user-supplied text (jokes). This creates a vulnerability surface where malicious instructions embedded within the text could attempt to override the agent's behavior during the diagnostic process. The skill does not include specific sanitization or boundary marker instructions for handling this untrusted input.
Audit Metadata