tolaria-wiki

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text can enter the LLM context via Raw note bodies created from external sources (e.g., Slack/email/ticket exports) in references/operations/integrations.md, then read during Raw processing / summary generation context packs in references/operations/maintenance.md and references/orchestration/automation.md.