subagents
Execution-layer skill for spawning subagents via runAgent
When & Why to Use This Skill
This Claude skill provides a robust execution-layer mechanism for spawning and managing subagents using the runAgent command. It is designed to streamline multi-agent orchestration, enabling a primary agent to delegate complex tasks to specialized sub-workers while maintaining strict governance and limited tool permissions for enhanced security and efficiency.
Use Cases
- Complex Task Decomposition: Efficiently break down large-scale projects into smaller, manageable units handled by specialized subagents to ensure high-quality outputs.
- Governance and Security: Execute sensitive operations by spawning subagents with restricted tool access, ensuring that sub-tasks are performed within a controlled and secure environment.
- Scalable Multi-Agent Workflows: Orchestrate parallel processing of diverse tasks by delegating to multiple subagent instances, significantly reducing total execution time for complex workflows.
| name | subagents |
|---|---|
| description | Execution-layer skill for spawning subagents via runAgent |
Commands
- runAgent
Constraints
- Subagents must run with explicit, limited tool permissions.
- Governance-layer routing decisions happen before runAgent is called.
Notes
- Governance decides which agent id to call; this skill is a thin wrapper around runAgent.