solo-swarm

Warn

Audited by Snyk on Apr 5, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 1.00). The skill instructs agents to perform web searches and scrape public, user-generated sources (e.g., site:reddit.com, news.ycombinator.com, site:github.com, and using WebFetch/WebSearch) and then synthesize those findings into research and GO/NO-GO recommendations, so untrusted third-party content can directly influence agent decisions.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 5, 2026, 09:51 AM
Issues
1