company-brain

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

• Third-party content exposure detected (high risk: 0.75). High likelihood: the required runtime workflow for query/compile reads the team vault’s structured raw markdown (e.g., meetings/, customer-language/, recurring-questions/) that can be auto-synced from outsider-authored sources like Slack/email/support tickets via toolify/webhooks, and those markdown bodies become LLM-readable context during compilation/query.