ai-content-pipeline

Warn

Audited by Snyk on May 28, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.75). The workflow’s “Research -> Content” pattern uses tavily/search-assistant, which at runtime fetches public web pages/snippets and feeds that retrieved free text into the downstream LLM context for summarization/content generation (public web content = outsider-authored text).

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
May 28, 2026, 11:38 AM
Issues
1
Security Audit — snyk — ai-content-pipeline