Template Skill
Bootstrap new Claude skills from a minimal, production-ready template that follows ClaudeKit skill conventions.
What This Skill Does
The Challenge: Every new skill needs the same boilerplate — SKILL.md format, directory structure, activation patterns, and configuration files. Starting from scratch wastes time and produces inconsistent skill quality.
The Solution: Template Skill provides a minimal but complete starting point for new skills. Covers essential sections, avoids over-engineering, and follows ClaudeKit conventions established across the skill library.
Activation
Explicit: Activate via prompt:
Activate template-skill to scaffold a new skill called "data-export"
Capabilities
1. Skill Directory Scaffold
Generate the minimum required file structure for a new skill.
Generated structure:
skills/your-skill-name/
├── SKILL.md # Required: skill definition
├── README.md # Optional: user-facing docs
└── examples/
└── basic-example.md # At least one example
2. SKILL.md Template
Minimal SKILL.md with required sections pre-filled.
# Your Skill Name
## Overview
One paragraph: what it does, when it activates, primary use case.
## Activation
**Implicit**: [conditions] OR "Does not auto-activate."
**Explicit**: `Activate your-skill-name to [task description]`
## Capabilities
### 1. Core Capability
Description with example.
## Configuration
List of required env vars or files. "None required." if applicable.
## Examples
### Basic Example
Input: [user prompt]
Output: [expected result description]
3. Naming Convention Guide
Rules for skill naming and file organization.
Naming rules:
- Use kebab-case:
social-media, notsocialMedia - Be descriptive:
email-sequence-builder, notemail - Match invocation: skill named
data-exportinvoked as/ckm:data-export
4. Activation Pattern Templates
Common activation patterns to copy.
Implicit (content-triggered):
**Implicit**: Activates when user asks to [describe trigger context].
Implicit (agent-triggered):
**Implicit**: Activates automatically for [Agent Name] agent.
Explicit only:
**Implicit**: Does not auto-activate.
**Explicit**: `Activate skill-name to [describe task]`
Prerequisites
- Skill name and primary purpose defined
- Target activation context decided (implicit/explicit)
Configuration
No configuration required for template itself. Generated skill may have its own requirements.
Best Practices
1. Keep SKILL.md under 100 lines Skills should be focused. Long SKILL.md files signal a skill doing too much.
2. One example per capability Each capability listed needs at least one concrete input/output example.
3. Be explicit about scope boundaries Include a “Does not” or “Out of scope” note if the skill name could be misinterpreted.
Common Use Cases
Use Case 1: First Skill Creation
Scenario: Developer creating their first ClaudeKit skill.
Workflow:
- Run template scaffold with skill name
- Fill in SKILL.md sections
- Add one concrete example
- Test activation with explicit prompt
- Refine based on output quality
Output: Working skill directory ready for testing.
Use Case 2: Skill Standardization
Scenario: Existing skills have inconsistent formatting across a skill library.
Workflow:
- Use template as reference for SKILL.md structure
- Audit each existing skill against template sections
- Add missing sections (examples, configuration)
- Standardize activation patterns
Output: Consistent skill library following shared conventions.
Troubleshooting
Issue: Skill not activating with expected prompt Solution: Review activation pattern. Explicit prompts must contain the skill name or explicit activation phrase.
Issue: Skill activates but produces wrong output format Solution: Add explicit output format instructions to SKILL.md capabilities section.
Related Skills
- Skill Creator - Full skill creation with benchmarks
- Docs - Documentation generation
Related Commands
/ckm:template-skill- Scaffold new skill from template/ckm:skill-creator- Create fully-specified skill with tests