task

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly reads task data from third-party backends (e.g., beans list --json, linear issue list, gh issue list and task bodies referenced in commands/analyse.md and update.md) and uses those user-generated task bodies to decide actions (e.g., whether to mark complete, include summaries, archive), exposing the agent to untrusted external content.