agent-management

Create and maintain Claude Code subagents with validated YAML frontmatter and phase-based task schemas.

1|Updated Nov 26, 2025
One-click install
npx skills add https://github.com/Emz1998/nexly-notes --skill agent-management
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-management
Source: https://github.com/Emz1998/nexly-notes/tree/main/.claude/skills/agents-management
Command: npx skills add https://github.com/Emz1998/nexly-notes --skill agent-management

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you proactively design, create, and maintain Claude Code subagents. It analyzes requirements, identifies specialized roles, crafts precise configurations with appropriate tool permissions and model tiers, and updates existing agents while preserving core functionality.

Core Features & Use Cases

  • Agent design and configuration: create or update agent persona, roles, responsibilities, and tasks.
  • Tool and model tuning: select minimal tool permissions and appropriate model tier for cost-effectiveness.
  • Change management: handle agent creation and updates with branching and validation, plus comprehensive reporting.
  • Use Case: When starting a new Claude Code project, quickly bootstrap optimal subagents and align them with project requirements, or update an existing agent to add new capabilities while keeping its core behavior intact.

Quick Start

Create or update a subagent named [agent-name] using the template at .claude/skills/agents-management/template.md, or update an existing one if [agent-name] already exists.

Frequently Asked Questions about agent-management

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

FAQPage Schema
How do I create Claude Code subagents for my project?

Create subagents by analyzing your project requirements, identifying specialized roles, and configuring agents using the provided template with appropriate tool permissions and model tiers. The Skill enforces YAML frontmatter validation and phase-based task schema to ensure compliance and minimal permissions.

What's the best way to select tool permissions and model tiers for a subagent?

Analyze your agent's core responsibilities and grant only the minimal tool permissions required for those tasks. Match the model tier to task complexity and cost constraints—lower tiers for simpler operations, higher tiers for complex reasoning—balancing capability with efficiency.

Can I update an existing subagent while preserving its core functionality?

Yes. The Skill handles agent updates through branching and validation, applying changes to existing agents while maintaining their core behavior. It enforces template compliance and controlled branch management to prevent unintended capability drift.

What naming conventions and responsibility restrictions apply to Claude Code subagents?

Subagents must follow strict naming standards and enforce restricted core responsibilities as defined by the phase-based task schema (T001-T012). This ensures consistency across projects and prevents scope creep in agent capabilities.

Do I need to bootstrap multiple subagents when starting a new Claude Code project?

The Skill accelerates bootstrapping by designing and configuring multiple optimal subagents aligned with your project requirements in a single workflow, eliminating manual role identification and configuration for each agent.

What validation does the agent-management Skill enforce?

The Skill validates YAML frontmatter, enforces the phase-based task schema (T001-T012), verifies strict naming conventions, ensures minimal tool permissions, and confirms template compliance across all agent configurations.