How do I install the Write A Skill skill?

Run npx skillmds@latest add zap-coding-agent/write-a-skill in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Write A Skill safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Write A Skill?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Write A Skill free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Write A Skill?

zap-coding-agent (@zap-coding-agent) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Write A Skill by zap-coding-agent — Agent Skill

Name: Write A Skill
Rating: 4.666666666666667 (3 reviews)
Author: zap-coding-agent

Write A Skill

Use when the user wants to create a new agent skill — gathers requirements, drafts SKILL.md with proper description format, adds reference files if needed, and reviews with the user. Triggers on "create a skill", "write a skill", "build a new skill", or "add a skill".

File contents

Write a Skill

Create a new agent skill with proper structure following the agentskills.io specification.

Process

1. Gather requirements

What task or domain does the skill cover?
What specific use cases and triggers should it handle?
Does it need reference files (command tables, templates, checklists)?
Any executable scripts for deterministic operations?

2. Draft the skill

skill-name/
├── SKILL.md           # Main instructions (required)
├── reference.md       # Optional: tables, commands, lookup content
└── template.md        # Optional: templates the skill fills out

3. Write the description correctly

The description field is how agents decide whether to load this skill. It must:

Start with "Use when..." — triggering conditions and symptoms only
Never summarize the workflow (agents follow the description instead of reading the body)
Include error messages, tool names, or command names that would appear in context
Be keyword-rich for discovery

# ✅ Correct
description: Use when writing Splunk SPL — authoring searches, fixing pipeline errors, choosing between stats/tstats, applying CIM field names, or the search returns unexpected results.

# ❌ Wrong — summarizes workflow
description: Guides you through writing optimized Splunk searches using best practices and CIM normalization.

4. Review with the user

Does this cover your use cases?
Anything missing or unclear?
Does the description trigger in the right situations?

Skill Body Guidelines

One excellent example over many mediocre ones — real, runnable
Heavy reference content → separate file (keeps SKILL.md scannable)
Cross-reference related skills with REQUIRED COMPANION: — never @-link
No workflow summaries in the description — that's what the body is for

REQUIRED COMPANION: pocock-write-a-skill — Matt Pocock's philosophy behind skill authoring.

---name: write-a-skilldescription: Use when the user wants to create a new agent skill — gathers requirements, drafts SKILL.md with proper description format, adds reference files if needed, and reviews with the user. Triggers on "create a skill", "write a skill", "build a new skill", or "add a skill".---# Write a SkillCreate a new agent skill with proper structure following the agentskills.io specification.## Process**1. Gather requirements**- What task or domain does the skill cover?- What specific use cases and triggers should it handle?- Does it need reference files (command tables, templates, checklists)?- Any executable scripts for deterministic operations?**2. Draft the skill**```skill-name/├── SKILL.md # Main instructions (required)├── reference.md # Optional: tables, commands, lookup content└── template.md # Optional: templates the skill fills out```**3. Write the description correctly**The `description` field is how agents decide whether to load this skill. It must:- Start with "Use when..." — triggering conditions and symptoms only- Never summarize the workflow (agents follow the description instead of reading the body)- Include error messages, tool names, or command names that would appear in context- Be keyword-rich for discovery```yaml# ✅ Correctdescription: Use when writing Splunk SPL — authoring searches, fixing pipeline errors, choosing between stats/tstats, applying CIM field names, or the search returns unexpected results.# ❌ Wrong — summarizes workflowdescription: Guides you through writing optimized Splunk searches using best practices and CIM normalization.```**4. Review with the user**- Does this cover your use cases?- Anything missing or unclear?- Does the description trigger in the right situations?## Skill Body Guidelines- **One excellent example** over many mediocre ones — real, runnable- **Heavy reference content** → separate file (keeps SKILL.md scannable)- **Cross-reference** related skills with `REQUIRED COMPANION:` — never `@`-link- **No workflow summaries in the description** — that's what the body is for**REQUIRED COMPANION:** pocock-write-a-skill — Matt Pocock's philosophy behind skill authoring.

Frequently asked questions