amazon-automation

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

• Third-party content exposure detected (high risk: 0.70). The required runtime path uses Rube MCP tool calls (RUBE_SEARCH_TOOLS, RUBE_MANAGE_CONNECTIONS, RUBE_MULTI_EXECUTE_TOOL) whose returned tool schemas/plans are vendor-provided text that the agent ingests, which is outsider-authored free text from a third-party MCP server.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires connecting at runtime to the Rube MCP endpoint (https://rube.app/mcp) to fetch tool schemas, recommended execution plans, and to execute Composio tools—remote content that directly controls tool selection/instructions and thus agent behavior.