etermin-automation

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly requires calling RUBE_SEARCH_TOOLS against the Rube MCP endpoint (e.g., https://rube.app/mcp and composio.dev/toolkits/etermin) to fetch live tool schemas, recommended execution plans, and arguments which the agent must read and act on to select tool slugs and invoke actions, so third-party responses can directly influence behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires connecting to the MCP server at https://rube.app/mcp and calling runtime endpoints (RUBE_SEARCH_TOOLS, RUBE_MANAGE_CONNECTIONS, RUBE_MULTI_EXECUTE_TOOL) which fetch tool schemas/recommended execution plans that directly determine agent prompts/behavior and can trigger remote tool execution.