scan-code-accessibility-figma

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 HTML from a runtime-supplied file path (readFileSync(opts.file, "utf8")) or inline --html string, then JSDOM parses it and runs window.axe.run(...), so any free text inside that HTML (e.g., comments/attributes) becomes part of the LLM-visible JSON output (e.g., nodes[].html and failureSummary).