mcporter-setup

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and inspects arbitrary MCP server URLs provided by the user (scripts/detect-auth.ts performs GET/POST to the supplied URL as described in SKILL.md Step 3) and enumerates/parses remote tool metadata via mcporter (scripts/test-server.ts / SKILL.md Step 6), so untrusted third-party content is ingested and can materially influence subsequent actions (which tools to call and auth flows).

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

• Potentially malicious external URL detected (high risk: 0.80). The skill performs runtime requests to MCP endpoints (for example https://mcp.linear.app/mcp) and invokes npx/mcporter at runtime, so remote MCP servers can supply tool metadata that influences agent behavior and npx can fetch/execute remote code.