create-subagent

Create and manage Cursor subagents with YAML frontmatter and system prompts.

Updated Apr 11, 2026
One-click install
npx skills add https://github.com/fredericpoitevin/.cursor --skill create-subagent-fredericpoitevin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-subagent
Source: https://github.com/fredericpoitevin/.cursor/tree/main/skills-cursor/create-subagent
Command: npx skills add https://github.com/fredericpoitevin/.cursor --skill create-subagent-fredericpoitevin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enables teams to design specialized AI subagents that run in isolated contexts to handle domain-specific tasks without polluting main conversations.

Core Features & Use Cases

  • Define project-level and user-level subagents using YAML frontmatter and a system-prompt body.
  • Store and reuse subagent configurations across projects and machines.
  • Quickly prototype, test, and deploy task-specific assistants for development, code review, debugging, or domain tasks.

Quick Start

Create a new subagent file under .cursor/agents or ~/.cursor/agents, add a YAML frontmatter with name and description, and place the system prompt in the body.

Frequently Asked Questions about create-subagent

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

FAQPage Schema
How do I create subagents in Cursor for specialized AI tasks?

To create subagents in Cursor, you add a .md file under .cursor/agents or ~/.cursor/agents with a YAML frontmatter containing name and description, using the body as the system prompt. This enables focused, isolated AI assistants for domain-specific tasks without polluting main conversations.

What is the difference between project-level and user-level subagents?

Project-level subagents are stored in .cursor/agents and apply only to that specific project, while user-level subagents are stored in ~/.cursor/agents and can be reused across projects and machines. Both use the same YAML frontmatter and system prompt configuration format.

How do I extend subagent behavior with scripts and references in Cursor?

You can extend subagent behavior by adding optional scripts/, references/, and assets/ directories alongside the .md configuration file. These directories provide supplementary resources that the subagent can leverage to handle specialized tasks like code review, debugging, or domain-specific workflows.

Does configuring Cursor subagents require any dependencies?

Configuring Cursor subagents requires no external dependencies. You only need a Markdown file with YAML frontmatter defining the name and description, plus the system prompt in the body. The configuration is self-contained and portable across environments.

When should I use isolated subagents instead of a single AI conversation?

You should use isolated subagents when you need specialized AI to handle domain-specific tasks without polluting main conversations. Subagents keep contexts separate, making them ideal for dedicated development, code review, debugging, or other focused workflows within Cursor.