worker-dispatch

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and injects GitHub issue bodies and comments (via the extract_issue_context function using gh issue view and the HANDOVER/PROGRESS fields) — user-generated, potentially public content — directly into worker prompts for interpretation, exposing the agent to untrusted third-party content.