manimce-best-practices

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

• Third-party content exposure detected (medium risk: 0.30). In examples/attention/scenes.py, phrase = "a fluffy blue creature roamed the verdant forest" is split into words and each word is rendered as Text(word, ...), so the workflow ingests runtime free text (user/outsider-authored if the scene input is changed) without selecting a specific trusted item first.