mermaid-studio

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime ingests user-supplied Mermaid text from the .mmd file content (or --stdin) during validation and rendering (e.g., scripts/validate.mjs reads the file/stdin, scripts/render.mjs reads the diagram file and passes its content to Mermaid renderers, including Puppeteer icon mode where mmdContent is embedded into HTML and rendered).

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

• Potentially malicious external URL detected (high risk: 0.90). The render script loads and executes remote JavaScript and icon JSON at runtime (e.g., importing mermaid from https://cdn.jsdelivr.net/npm/mermaid@11/dist/mermaid.esm.min.mjs and fetching icon packs from https://unpkg.com/@iconify-json/... ), which causes execution of remote code/data during rendering when icon-enabled rendering is used.