bright-data
Warn
Audited by Snyk on Apr 17, 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.md explicitly instructs using Bright Data to scrape open/public social media and websites (e.g., the "Social Media Scraping" and "How to Use" sections with examples POSTing arbitrary Twitter/Reddit/YouTube/Instagram URLs to Bright Data and consuming returned posts/comments/transcripts), which ingests untrusted user-generated content that the agent is expected to read and could materially influence subsequent actions.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata