Skip to content
Back to skills

Add Skill

ASecurity

Guide for adding a new skill to an existing plugin. Use when creating or adding skills to plugins in this marketplace.

  • 57 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added October 17, 2026
ai-agentsbash

Security analysis

A100/100

Scanned October 17, 2026

npx -y skills add diegosouzapw/awesome-omni-skill --skill add-skill --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Add Skill?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Add Skill
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/diegosouzapw-add-skill/badge)](https://www.skillsdirectory.com/skills/diegosouzapw-add-skill)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: add-skill
description: Guide for adding a new skill to an existing plugin. Use when creating or adding skills to plugins in this marketplace.
---

# Add Skill Guide

Guide for adding a new skill to an existing plugin in this marketplace.

## Skill Directory Structure

```
plugins/<plugin-name>/skills/<skill-name>/
├── SKILL.md           # Required: skill definition
├── *.md               # Optional: additional instructions
└── scripts/           # Optional: executable scripts
    └── *.sh
```

## SKILL.md Structure

Every skill requires a `SKILL.md` file with YAML frontmatter:

```markdown
---
name: skill-name
description: Brief description of what this skill does and when to use it
---

# Skill Name

## Instructions

[Clear, step-by-step guidance for Claude to follow]

## Examples

[Concrete examples of using this skill]
```

## Required Fields

### name

- Maximum 64 characters
- Lowercase letters, numbers, and hyphens only
- Cannot contain "anthropic" or "claude"

### description

- Must be non-empty
- Maximum 1024 characters
- Include both what the skill does AND when to use it

## Three Levels of Content

### Level 1: Metadata (always loaded)

YAML frontmatter with `name` and `description`. Loaded at startup (~100 tokens).

### Level 2: Instructions (loaded when triggered)

Main body of SKILL.md. Loaded when skill is triggered (under 5k tokens
recommended).

### Level 3: Resources (loaded as needed)

Additional files and scripts. Only loaded when referenced.

## Using Scripts

For complex logic, create executable scripts:

```
skills/<skill-name>/scripts/
└── do-something.sh
```

Benefits:

- Script code never enters context window
- Only script output consumes tokens
- More reliable than generated code

## Script Path Resolution

**Important**: Skills are loaded from a plugin cache, not the repository.
Hardcoded paths like `plugins/foo/skills/bar/scripts/script.sh` will NOT work.

When referencing scripts in SKILL.md, use this pattern:

```markdown
**Script location**: `scripts/my-script.sh` (relative to this skill's directory)

Before running, locate this skill's directory (where this SKILL.md is located),
then execute: bash <skill-directory>/scripts/my-script.sh
```

This ensures Claude finds the correct script path regardless of where the plugin
is installed.

## Reference

- Agent Skills overview:
  https://platform.claude.com/docs/en/agents-and-tools/agent-skills/overview
- Best practices:
  https://platform.claude.com/docs/en/agents-and-tools/agent-skills/best-practices

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.