architecture-review-gate

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

• Third-party content exposure detected (medium risk: 0.30). The only runtime text ingestion is the user-supplied architecture Markdown file passed to scripts/score_architecture.py (read via read_text()), i.e., outsider-authored free text can be submitted by providing that file, and the workflow then scans that text without selecting a specific item first.