ru-score
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted text input via the
$ARGUMENTSvariable, creating a surface for indirect prompt injection where malicious instructions could attempt to influence the scoring logic. - Ingestion points: External text provided by the user in
$ARGUMENTSor extracted from recent output (SKILL.md). - Boundary markers: Absent. The instructions do not specify delimiters or isolation markers to separate the input text from the scoring instructions.
- Capability inventory: The skill explicitly restricts its toolset to read-only capabilities (
Read,Grep,Glob) and disallows dangerous tools such asBash,PowerShell, andWrite(SKILL.md), which serves as a significant mitigation. - Sanitization: Absent. No filtering or validation is applied to the analyzed text before processing.
Audit Metadata