Skip to content
Back to skills

Create Skill

ASecurity

Create a new SideCar skill file with proper frontmatter, structure, and best practices

  • 8 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added October 4, 2026
ai-agentsgit

Security analysis

A100/100

Pro scans all 21 files and shows the line behind each finding

Scanned October 4, 2026

npx -y skills add nedonatelli/sidecar --skill skills --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Create Skill?

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

Security grade badge for Create Skill
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/nedonatelli-create-skill/badge)](https://www.skillsdirectory.com/skills/nedonatelli-create-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: Create Skill
description: Create a new SideCar skill file with proper frontmatter, structure, and best practices
---

# Create a New SideCar Skill

The user wants to create a new skill. Guide them through the process and generate the skill file.

## Skill File Format

Skills are Markdown files with YAML frontmatter. They live in one of these directories:

- `~/.claude/commands/` — personal skills (available in all projects)
- `<workspace>/.claude/commands/` — project-specific skills (shared via git)
- `<workspace>/.sidecar/skills/` — SideCar native project skills

## Required Structure

```markdown
---
name: Human-Friendly Name (max 64 chars)
description: When and how this skill should be used (max 200 chars)
---

# Skill Title

Instructions for the AI agent when this skill is invoked.
```

## Writing Effective Skills

1. **Focus on one task** — create separate skills for different workflows
2. **Be specific in the description** — SideCar uses the description to match skills to user queries
3. **Include context** — explain what files, patterns, or conventions to follow
4. **Add examples** — show sample inputs and expected outputs
5. **Use imperative instructions** — tell the agent what to DO, not what it IS

## Steps

1. Ask the user what the skill should do
2. Ask where it should be saved (personal `~/.claude/commands/` or project `.sidecar/skills/`)
3. Generate the skill file with:
   - A clear, concise `name` (max 64 chars)
   - A `description` that explains when to trigger it (max 200 chars)
   - Structured markdown body with step-by-step instructions
4. Write the file using `write_file`
5. Confirm the skill was created and explain how to use it (`/skill-id` in chat)
6. If the user has `sidecar.skills.userRegistry` configured, ask: "Would you like to publish this skill to your registry so it's available on all your machines?" If yes, tell them to run `SideCar: Publish Skill to Registry` from the Command Palette with the new file open — you cannot invoke VS Code commands yourself.

## Naming Convention

The filename (without `.md`) becomes the slash command. Use kebab-case:

- `fix-tests.md` → `/fix-tests`
- `add-component.md` → `/add-component`
- `review-pr.md` → `/review-pr`

Files in this skill

  • agent-dx-cli-scale.md3.8 KB
  • break-this.md2.2 KB
  • create-skill.md2.2 KB
  • debug.md1.5 KB
  • explain-code.md1.4 KB
  • fabric/agility_story.md885 B
  • fabric/analyze_email_headers.md2.6 KB
  • fabric/analyze_incident.md1.8 KB
  • fabric/analyze_logs.md2.3 KB
  • fabric/analyze_malware.md3 KB
  • fabric/analyze_paper.md6.1 KB
  • fabric/analyze_paper_simple.md7 KB
  • fabric/analyze_tech_impact.md2 KB
  • fabric/analyze_terraform_plan.md1.1 KB
  • fabric/analyze_threat_report.md2.3 KB
  • fabric/ask_secure_by_design_questions.md2.1 KB
  • fabric/clean_text.md609 B
  • fabric/coding_master.md1.3 KB
  • fabric/compare_and_contrast.md328 B
  • fabric/convert_to_markdown.md1.7 KB

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…