homeowner-question-content

Warn

Audited by Snyk on Jul 3, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.95). The required runtime workflow explicitly calls UnifAPI to fetch live public web/Reddit/news content (e.g., seo/serpreddit/posts/{id}/comments and news/search), which is outsider-authored free text that the agent reads and incorporates into its LLM context.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 3, 2026, 04:06 PM
Issues
1
Security Audit — snyk — homeowner-question-content