create-subagents

Create and configure Claude Code subagents with structured XML prompts.

5|3|Updated Jan 6, 2026
One-click install
npx skills add https://github.com/durandom/skills --skill create-subagents-durandom
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-subagents
Source: https://github.com/durandom/skills/tree/main/.claude/skills/create-subagents
Command: npx skills add https://github.com/durandom/skills --skill create-subagents-durandom

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Create Subagents skill provides a structured blueprint for crafting specialized Claude Code subagents (and Task tool workflows). It helps teams formalize agent roles, tools, and constraints, enabling consistent orchestration and reuse across projects.

Core Features & Use Cases

  • Defining role, focus areas, workflow, constraints for subagents to ensure deterministic behavior
  • Generating project-level and user-level subagents, with clear precedence and handoff rules
  • Guidance for tool access, memory considerations, and evaluation strategies
  • Use cases: rapidly spin up domain-specific subagents for code reviews, testing, security analysis, or automation tasks
  • Optional: reuse patterns in large multi-agent systems to orchestrate tasks

Quick Start

  • Create a new subagent by following the SKILL.md format and placing it under .claude/agents/
  • Use /agents to manage and test subagents, then invoke them via the main workflow
  • Example: create a stateless subagent for code-reading tasks

Frequently Asked Questions about create-subagents

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create Claude Code subagents for multi-agent orchestration?

To create Claude Code subagents, define role, focus areas, constraints, and workflow in a structured XML prompt format, then place the SKILL.md file under .claude/agents/. This enforces deterministic behavior and safe tool access patterns for reliable orchestration.

What is the structured XML prompt format for Claude Code subagents?

The structured XML prompt format enforces strict role definitions, constraints, and workflow rules for Claude Code subagents. It ensures consistent, auditable behavior by standardizing memory guidelines and evaluation hooks across project-level and user-level scopes.

How do I configure tool access and memory for Claude Code subagents?

Configuring tool access and memory involves defining safe tool access patterns and memory guidelines within the subagent's XML prompt. This ensures stateless or stateful behavior is controlled, enabling reliable task execution and consistent multi-agent handoffs.

Does this subagent orchestration approach support project-level and user-level scopes?

Yes, subagent orchestration supports both project-level and user-level scopes with clear precedence and handoff rules. You can generate domain-specific subagents for code reviews or security analysis and reuse patterns across large multi-agent systems.

What are the limitations of managing Claude Code subagents?

Limitations include ensuring strict role and constraint definitions are maintained to prevent non-deterministic behavior. Subagents require evaluation hooks to audit behavior, and complex multi-agent orchestration needs careful tool access configuration to avoid conflicts.