tool-design

1 stars 2 forks
17
A

Design tools that agents can use effectively. Use when creating new tools for agents, debugging tool-related failures, or optimizing existing tool sets.

Third-Party Agent Skill: Review the code before installing. Agent skills execute in your AI assistant's environment and can access your files. Learn more about security

Installation for Agentic Skill

View all platforms →
skilz install mjunaidca/mjs-agent-skills/tool-design
skilz install mjunaidca/mjs-agent-skills/tool-design --agent opencode
skilz install mjunaidca/mjs-agent-skills/tool-design --agent codex
skilz install mjunaidca/mjs-agent-skills/tool-design --agent gemini

First time? Install Skilz: pip install skilz

Works with 22+ AI coding assistants

Cursor, Aider, Copilot, Windsurf, Qwen, Kimi, and more...

View All Agents
Download Agent Skill ZIP

Extract and copy to ~/.claude/skills/ then restart Claude Desktop

1. Clone the repository:
git clone https://github.com/mjunaidca/mjs-agent-skills
2. Copy the agent skill directory:
cp -r mjs-agent-skills/.claude/skills/tool-design ~/.claude/skills/

Need detailed installation help? Check our platform-specific guides:

Related Agentic Skills

Agentic Skill Details

Stars
1
Forks
2
Type
Non-Technical
Meta-Domain
development
Primary Domain
javascript
Market Score
17

Agent Skill Grade

A
Score: 92/100 Click to see breakdown

Score Breakdown

Spec Compliance
12/15
PDA Architecture
26/30
Ease of Use
22/25
Writing Style
9/10
Utility
18/20
Modifiers: +5

Areas to Improve

  • Missing TOC in long files
  • Second-person voice in examples
  • Redundant 'Integration' section

Recommendations

  • Add trigger phrases to description for discoverability
  • Add table of contents for files over 100 lines

Graded: 2026-01-24

Developer Feedback

I've been digging into how tools evolve as systems grow more complex, and your skill nails the progression from "what should this do" to "how do we actually build it"—the fact that you scored 92 says you're handling the hard part most people skip over.

Links:

The TL;DR

You're at 92/100, solidly in A territory. This is based on Anthropic's best practices for skill design. Your writing style is the strongest pillar (9/10)—clear, objective, and technically sound. The weakest spot is spec compliance (12/15), mostly because your description could lean harder on trigger phrases for discoverability.

What's Working Well

  • Writing Style is tight — Your 9/10 here reflects genuinely professional docs. The consolidation principle is explained without fluff, and you use imperative language cleanly. The Tool Selection Framework feels like something someone would actually follow.

  • Architecture is thoughtfully layered — SKILL.md gives the overview, best_practices.md goes deep (177 lines), but you're not repeating yourself. That's how you score 26/30 on Progressive Disclosure Architecture. The reference file pattern lets someone get oriented quickly without drowning in details.

  • Utility is solid (18/20) — Your Tool-Testing Agent Pattern and the well-designed vs. poorly-designed tool comparisons give developers real leverage. The feedback loops section actually walks through how to iterate, not just talk about it.

The Big One: Missing Table of Contents

Here's what's holding back an extra 1-2 points: SKILL.md runs 270 lines and best_practices.md hits 177 lines, but neither has a TOC. This hurts navigation signals in your PDA score.

For someone scrolling through, a simple TOC after your metadata makes a h...

Report Security Issue

Found a security vulnerability in this agent skill?