miro-way-of-the-code-implement

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

• Third-party content exposure detected (high risk: 0.85). The skill ingests outsider-authored free text from a user-supplied Miro board at runtime via doc_get (reading each section’s H1 line and capturing doc_body), and also via diagram_get_dsl for the Mermaid diagrams; this content is not authored by the operating user and is fed into the agent’s LLM context for diffing and PR description generation.