vero-automation

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

• Third-party content exposure detected (high risk: 0.80). The skill requires calling RUBE_SEARCH_TOOLS against the public Rube MCP endpoint (https://rube.app/mcp) and ingesting returned tool schemas, recommended execution plans, and known pitfalls as part of its required workflow (see SKILL.md "Always search first" / Tool Discovery), so those public, dynamic responses could contain untrusted instructions that materially influence tool selection and execution.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires adding and calling the MCP endpoint https://rube.app/mcp at runtime (via RUBE_SEARCH_TOOLS / RUBE_MANAGE_CONNECTIONS) to fetch tool schemas and recommended execution plans that directly drive agent instructions and execution, so this external endpoint is a runtime dependency that can control prompts.