llm-wiki-capture-pipeline

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow (“capture -> inbox/raw -> wiki-triage -> wiki-ingest”) is explicitly designed to ingest outsider-authored content first into user inbox/raw material via named channels like email/Telegram/Slack/GitHub and then process it in subsequent triage/ingest steps, which enables indirect prompt injection through submitted messages/posts/issue text.