bmad-agent-marketing-community

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md "Starting Context" explicitly instructs the agent to "Audit the live community first" from a Live Community URL, meaning the agent will fetch and interpret public, user-generated community content (Discord/Slack/Reddit/forums) as a working source of truth and use it to drive recommendations and actions.