skill-parallel-agents

SUSPICIOUS. The skill is mostly coherent with its stated orchestration purpose and uses official install sources, so it does not look malicious. However, it materially increases risk by forcing automatic multi-provider execution, sharing task/context data with external LLM services, optionally routing through OpenRouter, and enabling AI-agent red-team/security-review workflows.