bot-developer

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

• Third-party content exposure detected (high risk: 0.70). The skill ingests and analyzes untrusted user-generated content — e.g., references/moderation-system.md's check_message reads message.content (extracting URLs, words, and expanding invites) and references/platform-templates.md shows receiving webhook updates — so it clearly processes third-party messages that could carry indirect prompt injection.