find-communities
Warn
Audited by Snyk on Apr 29, 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's "Enumerate" procedure explicitly uses "WebSearch + WebFetch" to scrape public community pages (e.g., LinkedIn groups, Slack/Discord listings, Reddit, mailing lists) and requires the agent to read activity/last-post signals and alignment from that untrusted third‑party content to decide which communities to recommend.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata