Skill template

Skill Pratiyush/agent-catalog/skills/skill/skill-template

Marketplace for AI Agent Skills, Prompts, Agents, and MCP servers. Powered by @skillscraft.

Install

npx -y skills add Pratiyush/agent-catalog --skill skill-template

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Scaffold new Agent Skills with best-practice structure and frontmatter. Use when the user wants to create a new skill, bootstrap a skill directory, or generate SKILL.md boilerplate following the Agent Skills spec.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.0 KB, as published. Nobody here has run it

Skill Template Generator

When to use this skill

Activate when the user wants to:

Create a new Agent Skill from scratch
Bootstrap a skill directory with proper structure
Generate SKILL.md boilerplate with valid frontmatter
Scaffold scripts/, references/, assets/, or tests/ subdirectories

Instructions

Quick scaffold

Ask the user for: skill name, short description, and which optional directories they need

Run the scaffold script:

node scripts/scaffold.js --name <skill-name> --desc "<description>" [--dirs scripts,references,assets,tests]

The script creates the full directory structure with template files
Walk the user through customising the generated SKILL.md body

Interactive mode

If the user is unsure what they need, ask these questions in order:

What does this skill do? → becomes the description field
Does it need executable scripts? → adds scripts/ with a starter script
Does it need reference docs? → adds references/ with a template
Does it need static assets? → adds assets/
Should it include test scenarios? → adds tests/scenarios.json
Which tools should agents use? → sets allowed-tools
Any runtime requirements? → sets compatibility
License? → defaults to MIT

Validating the result

After scaffolding, always validate:

skill validate <output-dir>/SKILL.md
skill lint <output-dir>/SKILL.md

Fix any issues before the user starts editing.

Name rules

Skill names must:

Be 1-64 characters
Match ^[a-z0-9]([a-z0-9-]*[a-z0-9])?$
Use lowercase letters, digits, and hyphens only
Not start or end with a hyphen

Valid: code-review, test-gen, my-skill-v2 Invalid: Code-Review, -bad-start, has_underscores, UPPERCASE

Output structure

<skill-name>/
  SKILL.md              # Always created
  scripts/              # If --dirs includes scripts
    run.sh              #   Starter bash script
  references/           # If --dirs includes references
    GUIDE.md            #   Template reference doc
  assets/               # If --dirs includes assets
    .gitkeep            #   Placeholder
  tests/                # If --dirs includes tests
    scenarios.json      #   Empty test scenarios array

Gotchas

Skill name must match the directory name — the script enforces this
Description must include "Use when" for best lint score
Keep SKILL.md body under 500 lines for optimal agent performance
The allowed-tools field accepts space-separated tool names (e.g., Bash Read Write)
If the user wants Python/Java/TS scripts, update the starter script language after scaffolding