Sub Agent Patterns
Automatically identifies recurring task sequences within agent workflows to optimize efficiency and suggest improvements.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add sub-agent-patterns npx -- -y @trustedskills/sub-agent-patterns
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"sub-agent-patterns": {
"command": "npx",
"args": [
"-y",
"@trustedskills/sub-agent-patterns"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
The sub-agent-patterns skill enables AI agents to delegate tasks to specialized sub-agents, improving efficiency and performance by breaking down complex workflows into smaller, focused operations. It supports dynamic task routing based on context or input, allowing for modular and scalable agent architectures.
When to use it
- When handling multi-step processes that require different expertise at each stage.
- To optimize resource usage in large-scale AI agent systems.
- For scenarios where tasks need to be parallelized or distributed across sub-agents.
- When building flexible, reusable components for agent-based applications.
Key capabilities
- Task delegation to specialized sub-agents
- Dynamic routing based on input context
- Modular and scalable architecture support
- Parallel processing of sub-tasks
Example prompts
- "Break down this customer service request into separate tasks and assign them to the appropriate sub-agents."
- "Use a sub-agent for data analysis and another for report generation."
- "Route this query to the finance sub-agent if it involves budgeting, or to the HR sub-agent otherwise."
Tips & gotchas
- Ensure that sub-agents are well-defined and have clear responsibilities to avoid overlap.
- Monitor performance when using multiple sub-agents in parallel to prevent bottlenecks.
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.