mcp-hub

Warn

Audited by Snyk on Mar 9, 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). This skill (SKILL.md) explicitly exposes MCP servers that connect Claude to external web resources—e.g., the "puppeteer" browser automation server and links to the MCP marketplace (mcp.run) and GitHub/Google Drive integrations—so the agent can fetch and read untrusted public webpages and user-generated content that could influence subsequent tool use.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 1.00). The skill's runtime configuration invokes npx to fetch and run remote MCP server packages (e.g., "npx -y @modelcontextprotocol/server-filesystem" and "npx -y @anthropic/mcp-google-drive"), which downloads and executes external code from the npm registry and therefore constitutes a runtime external dependency that executes remote code.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 9, 2026, 08:22 AM