Create Subagent
Dynamically generates specialized sub-agents to handle complex tasks, breaking down goals into manageable steps.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add create-subagent npx -- -y @trustedskills/create-subagent
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"create-subagent": {
"command": "npx",
"args": [
"-y",
"@trustedskills/create-subagent"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
The create-subagent skill allows an AI agent to dynamically generate and manage new, specialized sub-agents. These sub-agents can be assigned specific tasks or roles within a larger workflow, enabling the parent agent to handle more complex operations by delegating parts of its workload. This promotes modularity and improves overall efficiency in multi-step processes.
When to use it
- Complex Task Decomposition: Break down large, multifaceted goals into smaller, manageable sub-tasks handled by specialized agents.
- Role Assignment: Create distinct agents for specific roles like "researcher," "writer," or "editor" within a project.
- Parallel Processing: Run multiple tasks concurrently by assigning them to different sub-agents.
- Dynamic Workflow Adjustment: Generate new sub-agents on the fly based on changing requirements or unexpected events.
Key capabilities
- Sub-agent creation
- Task assignment to sub-agents
- Sub-agent management (start, stop)
- Workflow delegation
Example prompts
- "Create a sub-agent named 'Researcher' with the task of finding information on quantum computing."
- "Generate a new agent called 'Writer' and assign it the task of drafting an email based on the research findings from the 'Researcher' agent."
- “Start the ‘Editor’ sub-agent to review the draft email.”
Tips & gotchas
The skill requires careful planning of sub-agent roles and responsibilities. Ensure clear communication protocols are established between the parent agent and its sub-agents for optimal performance.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.