moscow
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- SAFE (SAFE): No malicious patterns or security risks detected across all files. The skill is entirely declarative and focuses on logical categorization tasks.\n- Indirect Prompt Injection (LOW): The skill processes user-provided lists via the
$ARGUMENTSplaceholder. While it lacks explicit boundary markers to separate instructions from data, the skill has no functional capabilities (no network access, file system access, or command execution). Any potential injection would be limited to manipulating the text output of the prioritization list, posing no threat to the host environment.
Audit Metadata