brightdata-scraper-studio

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow ingests outsider-authored free text when it builds/heals scrapers based on plain-English descriptions provided to the Bright Data CLI (e.g., bdata scraper create/run/heal ... "<what to extract...>" and bdata scraper heal ... "<what broke>" --url <url>), so an outsider can submit that free text into the monitored inputs the tool consumes.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The README instructs cloning the remote repository https://github.com/IncomeStreamSurfer/brightdata-scraper-studio-skill.git which fetches the SKILL.md (the agent skill/prompts) from that URL and thus the external repo content directly controls the agent's prompts at runtime.