mcp-builder
Warn
Audited by Snyk on Feb 16, 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 explicitly tells the agent to fetch public web resources (e.g., "Use WebFetch to load: https://modelcontextprotocol.io/llms-full.txt" and raw.githubusercontent.com SDK README URLs), meaning the agent will ingest and interpret untrusted third-party web content.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs using WebFetch at runtime to load the MCP protocol document which would directly inform/drive agent prompts/instructions (https://modelcontextprotocol.io/llms-full.txt), making it a required runtime dependency.
Audit Metadata