Create New Agent Skill

VerifiedSafe

Guide to creating specialized agent skills with progressive disclosure. Covers skill structure, creation steps, and best practices.

Sby Skills Guide Bot
DevelopmentIntermediate
7/22/2026
Claude Code
#skill-authoring#scaffolding#meta#progressive-disclosure#claude-code

Recommended for

Automated summary

This skill walks through authoring a new Agent Skill: gathering requirements, laying out the folder, generating a YAML-frontmattered SKILL.md, then testing the result.

Strengths

  • Clear five-step flow from the initial requirements interview through to verifying the finished skill
  • Ships a complete SKILL.md template with standard sections (When to Use, Prerequisites, Instructions, Verification)
  • Explains progressive disclosure and how to split content between SKILL.md and separate reference files
  • Includes a best-practices checklist and a final report format with improvement suggestions

Limitations

  • No validation or linting tooling: the quality of the generated SKILL.md rests entirely on hand-written content
  • Testing is only a manual simulation, with no automated run or real trigger verification
  • The template enforces sections (Prerequisites, Verification) that can be overkill for a very simple skill
When to use it

Use it when creating a brand-new reusable skill from scratch, including its folder layout and SKILL.md.

When not to use it

Skip it when you only need to fix, audit, or restructure an existing skill, since the flow is built around creation.

Security analysis

Safe
Quality score78/100

The skill only provides instructions for creating new skill files and directories. It does not declare or use any tools, and it contains no commands, network access, destructive actions, or data exfiltration. The only file operations are described as part of the skill authoring workflow, not executed by this skill itself.

No concerns found

How this analysis is made

Examples

Scaffold a new skill
Create a new skill called 'api-testing' that helps me write and run pytest-based API tests for a FastAPI service.
Skill with a helper script
/new-skill pdf-forms Extract and fill PDF form fields, and add a helper script for the field-parsing logic.
Skill with clear trigger conditions
I want a skill named 'release-notes' that activates whenever I ask to draft changelog entries from merged pull requests. Scaffold it for me.

Create a New Agent Skill

Create a new Agent Skill for specialized tasks.

Arguments

$ARGUMENTS should contain: <skill-name> [description]

What is a Skill?

A Skill is a folder containing:

  • SKILL.md - Main instructions with YAML frontmatter
  • Optional scripts and resources
  • Optional reference documentation

Skills use progressive disclosure: only the name and description are loaded initially. The full skill content is loaded when relevant.

Skill Creation Process

Step 1: Gather Information

Ask the user:

  1. Skill name: Short, descriptive (e.g., "pdf-forms", "api-testing")
  2. Purpose: What task does this skill help with?
  3. When to use: What triggers should activate this skill?
  4. Required tools: Any external dependencies?

Step 2: Create Skill Directory

skills/<skill-name>/
├── SKILL.md           # Main skill file
├── scripts/           # Optional helper scripts
│   └── helper.py
└── resources/         # Optional templates/examples
    └── template.md

Step 3: Generate SKILL.md

Create with this structure:

---
name: <skill-name>
description: <brief one-line description for discovery>
---

# <Skill Name>

## When to Use This Skill

<Clear description of when Claude should activate this skill>

## Prerequisites

<List any required tools, dependencies, or setup>

## Instructions

### Step 1: <First Step>
<Detailed instructions>

### Step 2: <Second Step>
<Detailed instructions>

...

## Reference Files

<Point to any additional documentation or scripts>

## Examples

### Example 1: <Common Use Case>
<Input/output examples>

## Verification

<How to verify the skill worked correctly>

Step 4: Progressive Disclosure Tips

Structure the SKILL.md for progressive disclosure:

  1. Top section: Most commonly needed info
  2. Middle sections: Detailed procedures
  3. Bottom sections: Edge cases and examples
  4. Separate files: Rarely-used reference material

Step 5: Test the Skill

  1. Simulate a task that should trigger the skill
  2. Verify Claude loads the skill appropriately
  3. Check that instructions are clear and complete
  4. Iterate based on observed behavior

Skill Best Practices

  1. Keep it focused: One skill = one capability
  2. Be specific: Clear trigger conditions
  3. Include verification: How to check success
  4. Use code wisely: Scripts can be tools OR documentation
  5. Think from Claude's perspective: What info does Claude need?

Output

Create the skill directory and files, then provide:

  • Summary of what was created
  • How to test the skill
  • Suggestions for improvement

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/create-new-agent-skill-mrwake1o/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/create-new-agent-skill-mrwake1o' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/create-new-agent-skill-mrwake1o/download' --output '.skill-downloads/create-new-agent-skill-mrwake1o/SKILL.md.part' && mv '.skill-downloads/create-new-agent-skill-mrwake1o/SKILL.md.part' '.skill-downloads/create-new-agent-skill-mrwake1o/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

This file does not provide a usable name/description header for this native procedure. It remains available to download and inspect; use the manual method below.

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/create-new-agent-skill-mrwake1o/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills