ai-slop-reviewer
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a robust security policy for handling PII (Personally Identifiable Information) as detailed in the referenced file
references/kr-pii-masking.md. It provides specific masking rules for Korean identifiers such as Resident Registration Numbers, Foreigner Registration Numbers, bank accounts, and addresses, aligning with South Korea's Personal Information Protection Act. - [PROMPT_INJECTION]: The skill functions as a text processor for user-provided content (blogs, reports, emails), which inherently presents an indirect injection surface. The risk is evaluated as safe because the skill is restricted to stylistic rewriting and does not utilize sensitive system tools, network capabilities, or file-writing operations. Ingestion points: User-supplied text provided for slop review in SKILL.md. Boundary markers: None explicitly defined for user input to distinguish it from instructions. Capability inventory: Text transformation and PII masking; no dangerous system tool access or network tools detected. Sanitization: Implements PII masking for specific Korean data patterns in
references/kr-pii-masking.md.
Audit Metadata