loom-enhanced-research

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

• Third-party content exposure detected (high risk: 0.85). The skill’s research loop can fetch and ingest evidence from outsider-authored public web content (e.g., via runtime search/dig actions using seed URLs/queries), and those fetched page/excerpt text is then captured into the LLM context for summarization and ledger/report generation.