ambiguity-report

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

• Third-party content exposure detected (medium risk: 0.30). The skill renders a user-supplied spec.json into output (via scripts/generate_site.py / scripts/generate_latex.py), and that spec contains outsider-authored free text fields like source.sections[].text, scenarios[].situation, weak_point, likely_outcome, redraft, and positions[].argument which are directly ingested at runtime for rendering.