skills.rest
dxas90
dxas90COMMUNITY
Explore all skills by this author →
Website𝕏@dxas90GitHub

agent-writer

Create and optimize Claude Code agent markdown files in .claude/agents/.

Updated Apr 7, 2026
Search Tags:#optimization#claude#agent#parallelization#creation#model-selection
One-click install
npx skills add https://github.com/dxas90/opsy-bag --skill agent-writer-dxas90
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-writer
Source: https://github.com/dxas90/opsy-bag/tree/main/.claude/skills/agent-writer
Command: npx skills add https://github.com/dxas90/opsy-bag --skill agent-writer-dxas90

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Teams often need to create, modify, and refine Claude agents, but doing so manually can lead to inconsistent documentation, suboptimal model choices, and missed parallelization opportunities.

Core Features & Use Cases

  • Creation Checklist: Guides users through purpose definition, tool selection, model selection, parallelization, error handling, and output formatting.
  • Model & Tool Guidance: Provides tables and recommendations for selecting the appropriate Claude model and necessary tools.
  • DRY Principle Enforcement: Prompts users to search existing agents before adding new content.
  • Parallelization Advice: Links to reference patterns to convert sequential tasks into parallel sub‑agents.
  • Optimization Checklist: Helps review and improve existing agents for performance and maintainability.
  • Use Cases: Creating a new code‑review agent, updating an existing deployment validator, or auditing agents for redundant logic.

Quick Start

Ask the agent-writer skill to create a new Claude agent for code review with the Sonnet model and the GitHub tool.

Frequently Asked Questions about agent-writer

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

FAQPage Schema
How do I create a Claude agent markdown file correctly?▼

To create a Claude agent markdown file, define its purpose, select appropriate tools, choose the correct Claude model, ensure DRY compliance, add error handling, and configure parallel execution patterns.

What is the best way to optimize existing Claude agents?▼

The best way to optimize existing Claude agents is to use a review checklist that evaluates tool selection, model choice, redundant logic, and opportunities to convert sequential tasks into parallel sub-agents.

Which Claude model should I select for my agent?▼

Selecting the appropriate Claude model depends on the agent's task complexity, with guidance tables available to match specific operational requirements like tool usage and parallelization with the optimal model.

How do I parallelize sequential tasks in Claude agents?▼

Parallelize sequential tasks by applying reference patterns that convert them into parallel sub-agents, ensuring the main agent markdown file includes correct parallel execution structures for improved performance.

Can I use this to update an existing deployment validator agent?▼

Yes, you can update an existing deployment validator by applying an optimization checklist that reviews documentation consistency, model choices, and error handling within the .claude/agents/ directory.

Related Skills

Explore More Software Engineering Skills→
Vivekmano27Vivekmano27

agent-builder

Generate Markdown agent definitions with frontmatter for Claude Code.

Community
Advanced
Folken2Folken2
2

claude-sdk-tool-creation

Create and register Python tools for Claude Agent SDK with MCP servers.

Community
Intermediate
michsindlingermichsindlinger
1

[PROJECT]-deployment-patterns

Configure CI/CD pipelines with Docker and secrets management.

Community
Advanced
tilework-techtilework-tech
149

Getting Started with Abilities

Enforce Read-tool usage, skill announcements, and TodoWrite checklists for AI workflows.

Official
Basic