create-subagents

Create and configure Claude Code subagents with XML system prompts.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Subagents enable scalable, autonomous workflows by encapsulating specialized roles, tool access, and orchestration logic for Claude Code.

Core Features & Use Cases

  • XML-based system prompts with pure XML structure
  • Fine-grained tool access and least-privilege controls
  • Multi-agent orchestration patterns (sequential, parallel, hierarchical)
  • Prompt caching and prompt-validation best practices
  • Use cases: creating, configuring, and governing subagents for coding, testing, security reviews, and automation tasks.

Quick Start

Use the /agents command to create or register subagents, or drop a new subagent into the project path .claude/agents/subagent-name.md. Define YAML frontmatter with name, description, tools, and model, and provide a pure XML system prompt in the body detailing the role, constraints, workflow, and evaluation rules.

Frequently Asked Questions about create-subagents

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

FAQPage Schema
How do I create and configure Claude Code subagents for automation workflows?

To create Claude Code subagents, define YAML frontmatter with name, description, tools, and model, then provide a pure XML system prompt in the body detailing role, constraints, workflow, and evaluation rules. Place the file in the project path .claude/agents/subagent-name.md.

What is multi-agent orchestration and how does it work with the Task tool?

Multi-agent orchestration coordinates specialized subagents using the Task tool through sequential, parallel, or hierarchical patterns. This approach encapsulates distinct roles and tool access to enable scalable, autonomous automation workflows within Claude Code.

How do I enforce least-privilege tool access controls for subagents?

Least-privilege tool access is enforced by defining specific allowed tools in the subagent's YAML frontmatter. This restricts the agent's capabilities to only the necessary functions, ensuring safe multi-agent workflows and strict governance.

Can I use XML prompts for Claude Code subagent system prompts?

Yes, Claude Code subagents require pure XML system prompts. The structured XML format enforces strict constraints, incorporates prompt caching practices, and follows best-practice validation for reliable role definition and workflow execution.

What are the limitations of using subagents for complex coding and security review tasks?

Subagents require rigid XML prompt structures and explicit tool access definitions to avoid misbehavior. Limitations include the need for strict governance and prompt-validation practices to ensure safe execution in complex automation, coding, and security review workflows.