mcp-builder
Warn
Audited by Snyk on Feb 21, 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). SKILL.md (Phase 1.2 and the "参考文件" entries) explicitly instructs the agent to WebFetch public URLs (e.g., https://modelcontextprotocol.io/sitemap.xml and https://raw.githubusercontent.com/.../README.md) and to read those third‑party pages as part of the workflow, so untrusted external content can directly influence tool use and agent actions.
Audit Metadata