signal-detection-pipeline
Warn
Audited by Snyk on Mar 14, 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). Yes — SKILL.md explicitly instructs the pipeline to run tools like "reddit-scraper" and "linkedin-post-research" (and to ingest event URLs, job postings, and web search results) which pull untrusted, user-generated public content that is then used to score leads and drive outreach decisions, allowing third-party content to influence agent actions.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata