text-human

Warn

Audited by Snyk on Aug 17, 2026

Risk Level: MEDIUM
Full Analysis

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

  • Third-party content exposure detected (medium risk: 0.30). The skill ingests outsider-authored free text when the required workflow is given an inline “text to humanize” that is inferred as the social/article flows (e.g., /brewtools:text-human review this reddit reply: "<text>"), and it then directly applies PASS 1/2 to that user-provided text without first selecting a trusted item.

Issues (1)

W011
MEDIUM

Third-party content exposure detected (indirect prompt injection risk).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 17, 2026, 01:36 AM
Issues
1
Security Audit — snyk — text-human