agentic-top-10
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: Evaluation of phrases such as 'ignore previous instructions' and 'admin mode' indicates they are part of a 'Prompt Injection Safety Notice'. These instructions are used defensively to guide the agent on how to handle malicious content found within analyzed code, rather than being an attempt to hijack the agent's own processing.
- [EXTERNAL_DOWNLOADS]: The skill provides a reference to an external repository on GitHub (github.com/fabraix/playground) as a resource for security testing. This reference is informational and does not initiate automated downloads or execute remote code.
Audit Metadata