miro-skill

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

• Third-party content exposure detected (high risk: 0.80). The skill reads and browses content from Miro boards (e.g., via get_items/get_board and the optional "miro-ai" plugin's "Browse board contents"), which are user-generated/third‑party items and can be publicly shared ("Anyone with link can view"), so the agent would ingest untrusted external content as part of its workflow.