mcp-architecture
Warn
Audited by Snyk on Jun 23, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill includes runtime commands that fetch and execute remote server packages (e.g., npx @modelcontextprotocol/server-github, docker run mcp-postgres-server, pip install mcp-server-slack) which install MCP servers from the Server Registry (https://github.com/modelcontextprotocol/servers) and thus can provide prompts/tools that directly control agent behavior.
Issues (1)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata