discord-reader

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly reads user-generated, public Discord content via opencli connected to the Discord Desktop (see SKILL.md Step 3 and references/commands.md "read" / "search" commands), and instructs using that content (e.g., "-f json" "for LLM context", summarization and sentiment) so untrusted third-party messages can be ingested and influence the agent's analysis or next actions.