inboxmate-demo

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

• Third-party content exposure detected (low risk: 0.10). The workflow scrapes external prospect websites using WebFetch to populate the chatbot's knowledge base.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill instructs the agent to make HTTP/JSON-RPC requests to an external MCP server at https://app.psquared.dev/api/mcp and use the OpenBrand API at https://openbrand.sh/api/extract at runtime using tokens from environment variables, establishing operational dependencies on third-party services.