ID8Labs Agent Suite
Warn
Audited by Snyk on Feb 16, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). The Scout (and related) skills explicitly mine community signals from Reddit, YouTube, and public forums and use MCPs like Firecrawl for competitor scraping and Perplexity for web research, which means the agent fetches and ingests untrusted, user-generated public web content as part of its workflow.
Audit Metadata