mermaid-to-proverif

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

• Third-party content exposure detected (high risk: 0.75). The required workflow is to parse a user-provided Mermaid sequenceDiagram (Step 1: “From the Mermaid diagram… Extract every participant or actor declaration” and Step 2: “Walk through every Note over annotation and message label”), so any free-text labels/notes in that diagram are outsider-authored input that can be injected into the LLM context during parsing/model generation.