serenity-stock-scorer

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

• Third-party content exposure detected (high risk: 0.75). The script reads t.text from the local SQLite snapshot by joining mentions→tweets and then performs keyword matching over that free-form tweet body (texts = [r['text'] ...]), so any outsider-authored tweet text present in the DB becomes LLM-readable context via the script’s JSON output (e.g., evidence[].signals and summary).