agent-swarm

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

• Third-party content exposure detected (medium risk: 0.30). In the required runtime workflow, the orchestrator “scouts the issue tracker” by reading outsider-authored issue bodies and full comment threads via gh api ...issues?.../glab api ...issues?... and then the workers “read the issue and its thread” via gh issue view <N> --comments/glab issue view <N> --comments, so attacker-controlled free text from issues/comments is ingested without pre-selecting a specific item beyond the tracker query.