encodian-automation

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

• Third-party content exposure detected (high risk: 0.70). The skill instructs the agent to call RUBE_SEARCH_TOOLS and RUBE_GET_TOOL_SCHEMAS against the external MCP server (https://rube.app/mcp / composio.dev) to fetch tool schemas and execution plans that the agent must read and interpret at runtime, exposing it to untrusted third-party content.