testing-anti-patterns
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- SAFE (SAFE): No malicious patterns, obfuscation, or security risks were detected across all 10 threat categories.
- NO_CODE (SAFE): The skill consists entirely of Markdown documentation and structural metadata. It does not include any source code, shell scripts, or binaries, eliminating the possibility of direct execution risks.
- PROMPT_INJECTION (SAFE): While the skill uses strong instructional language such as 'NEVER' and 'STOP', these are strictly confined to the context of software testing methodologies and do not attempt to override the AI agent's security boundaries or system prompts.
Audit Metadata