miro-automation

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow is driven by Composio/Rube MCP tool calls (e.g., MIRO_GET_BOARD_ITEMS, MIRO_GET_BOARDS2) whose responses can include outsider-authored free text from Miro boards (e.g., sticky note text/content) that the agent will read/parse into its LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly directs the agent to use the Rube MCP endpoint https://rube.app/mcp at runtime (see RUBE_SEARCH_TOOLS / RUBE_MANAGE_CONNECTIONS) to fetch tool schemas that directly influence agent behavior and instructions.