discord-bot-architect

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

• Third-party content exposure detected (high risk: 0.80). The skill processes untrusted, user-generated Discord content as part of its runtime (e.g., the Pycord slash command greet accepts a user-provided "message" option and the Discord.js interaction component collector handles arbitrary user responses), which exposes the agent to third-party inputs that could carry indirect prompt injections.