interview-prep
Warn
Audited by Snyk on Apr 19, 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 running TinyFish agents to scrape live public, user-generated content from Glassdoor, Blind, and Reddit (see "Step 2 — Parallel scraping" agent commands), so the agent will ingest and act on untrusted third‑party reports that can influence its outputs.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata