research-brief

Warn

Audited by Snyk on Jul 18, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.95). SKILL.md requires “Built-in web search + fetch” and “Fetch and read before citing,” meaning the agent will open outsider-authored public web pages at runtime and ingest their readable text into the LLM context (explicitly treated as untrusted input).

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 18, 2026, 06:05 AM
Issues
1
Security Audit — snyk — research-brief