duoduo-tether

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

• Third-party content exposure detected (medium risk: 0.30). The workflow connects external AI assistants to Duoduo via MCP and ingests messages/mail from them via endpoints and webhooks, allowing outsider-authored text to reach the agent.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill executes duoduo channel install @openduo/channel-tether which pulls an external package from an npm registry via URL/package locator at runtime.