ln-521-test-researcher
Warn
Audited by Snyk on May 11, 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). The SKILL.md workflow explicitly mandates using Web Search/WebFetch and MCP tools (Phase 3 & Phase 5) and calls out scraping public forums and user-generated sites (StackOverflow, Reddit) as required research sources, so the agent will ingest untrusted third‑party content from the open web that can materially influence its test recommendations and next actions.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata