slides-video

Warn

Audited by Snyk on May 13, 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's Step 2 "资料研究" (references/research-method.md) explicitly requires using WebFetch/WebSearch and browser automation to pull tweets, announcements, blogs, news, and read PDFs from public websites — content the agent must read and use to drive planning and generate PPTs/scripts, so it ingests untrusted third‑party material that could carry indirect prompt injections.

Issues (1)

W011
MEDIUM

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

Audit Metadata
Risk Level
MEDIUM
Analyzed
May 13, 2026, 09:11 PM
Issues
1