monday-automation

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

• Third-party content exposure detected (low risk: 0.10). Exclusion: The required runtime workflow only operates on Monday.com data via explicit tool calls (e.g., listing boards/items, then updating columns/comments), and the specification does not describe ingesting arbitrary outsider-authored free text from an external feed/queue without the agent first selecting a specific target item on the user’s behalf.

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

• Potentially malicious external URL detected (high risk: 0.90). This skill requires adding and contacting the MCP server at https://rube.app/mcp at runtime (e.g., call RUBE_SEARCH_TOOLS / verify the MCP responds), which means remote tool schemas from that URL will be fetched and can directly control the agent's available tools/instructions.