platform-norm-profiler

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

• Third-party content exposure detected (high risk: 0.72). Outsider free text can enter via “official platform docs fetched keyless” (public web pages scraped by scripts/connectors/firecrawl.py / tavily.py) and via “user-pasted policy text / user-provided exports/screenshots” that may include untrusted prose; this text is then read and incorporated into the LLM context to label/update norm-card rows.