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
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 17, 2026, 04:43 PM
Issues
1