create-skill

Name: create-skill
Author: NeoLabHQ

1.3kstars132forks

How to install create-skill?

neolabhq/context-engineering-kit/create-skill

$npx -y skills add NeoLabHQ/context-engineering-kit --skill create-skill

Installs into the current project.

Summary

Guide for creating effective skills. This command should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations. Use when creating new skills, editing existing skil

Files · 1

View on GitHub

SKILL.md

1	# Create Skill Command
2
3	This command provides guidance for creating effective skills.
4
5	## Overview
6
7	Writing skills IS Test-Driven Development applied to process documentation.
8
9	Personal skills live in agent-specific directories (`~/.claude/skills` for Claude Code, `~/.codex/skills` for Codex)
10
11	You write test cases (pressure scenarios with subagents), watch them fail (baseline behavior), write the skill (documentation), watch tests pass (agents comply), and refactor (close loopholes).
12
13	Core principle: If you didn't watch an agent fail without the skill, you don't know if the skill teaches the right thing.
14
15	REQUIRED BACKGROUND: You MUST understand Test-Driven Development before using this skill. That skill defines the fundamental RED-GREEN-REFACTOR cycle. This skill adapts TDD to documentation.
16
17	Official guidance: The Anthropic's official skill authoring best practices provided at the `/apply-anthropic-skill-best-practices` command, they enhance `prompt-engineering` skill. Use skill and the document, as they not copy but add to each other. These document provides additional patterns and guidelines that complement the TDD-focused approach in this skill.
18
19	## About Skills
20
21	Skills are modular, self-contained packages that extend Claude's capabilities by providing
22	specialized knowledge, workflows, and tools. Think of them as "onboarding guides" for specific
23	domains or tasks—they transform Claude from a general-purpose agent into a specialized agent
24	equipped with procedural knowledge that no model can fully possess.
25
26	## What is a Skill?
27
28	A skill is a reference guide for proven techniques, patterns, or tools. Skills help future Claude instances find and apply effective approaches.
29
30	Skills are: Reusable techniques, patterns, tools, reference guides
31
32	Skills are NOT: Narratives about how you solved a problem once
33
34	### What Skills Provide
35
36	1. Specialized workflows - Multi-step procedures for specific domains
37	2. Tool integrations - Instructions for working with specific file formats or APIs
38	3. Domain expertise - Company-specific knowledge, schemas, business logic
39	4. Bundled resources - Scripts, references, and assets for complex and repetitive tasks
40
41	## TDD Mapping for Skills
42
43	\| TDD Concept \| Skill Creation \|
44	\|-------------\|----------------\|
45	\| Test case \| Pressure scenario with subagent \|
46	\| Production code \| Skill document (SKILL.md) \|
47	\| Test fails (RED) \| Agent violates rule without skill (baseline) \|
48	\| Test passes (GREEN) \| Agent complies with skill present \|
49	\| Refactor \| Close loopholes while maintaining compliance \|
50	\| Write test first \| Run baseline scenario BEFORE writing skill \|
51	\| Watch it fail \| Document exact rationalizations agent uses \|
52	\| Minimal code \| Write skill addressing those specific violations \|
53	\| Watch it pass \| Verify agent now complies \|
54	\| Refactor cycle \| Find new rationalizations → plug → re-verify \|
55
56	The entire skill creation process follows RED-GREEN-REFACTOR.
57
58	## When to Create a Skill
59
60	Create when:
61
62	- Technique wasn't intuitively obvious to you
63	- You'd reference this again across projects
64	- Pattern applies broadly (not project-specific)
65	- Others would benefit
66
67	Don't create for:
68
69	- One-off solutions
70	- Standard practices well-documented elsewhere
71	- Project-specific conventions (put in CLAUDE.md)
72
73	## Skill Types
74
75	### Technique
76
77	Concrete method with steps to follow (condition-based-waiting, root-cause-tracing)
78
79	### Pattern
80
81	Way of thinking about problems (flatten-with-flags, test-invariants)
82
83	### Reference
84
85	API docs, syntax guides, tool documentation (office docs)
86
87	## Directory Structure
88
89	```
90	skills/
91	skill-name/
92	SKILL.md # Main reference (required)
93	supporting-file.* # Only if needed
94	```
95
96	Flat namespace - all skills in one searchable namespace
97
98	Separate files for:
99
100	1. Heavy reference (100+ lines) - API docs, comprehensive syntax
101	2. Reusable tools - Scripts, utilities, templates
102
103	Keep inline:
104
105	- Principles and concepts
106	- Code patterns (< 50 lines)
107	- Everything else
108
109	## Anatomy of a Skill
110
111	Every skill consists of a required SKILL.md file and optional bundled resources:
112
113	```
114	skill-name/
115	├── SKILL.md (required)
116	│ ├── YAML frontmatter metadata (required)
117	│ │ ├── name: (required)
118	│ │ └── description: (required)
119	│ └── Markdown instructions (required)
120	└── Bundled Resources (optional)
121	├── scripts/ - Executable code (Python/Bash/etc.)
122	├── references/ - Docu

People who liked this also viewed

6 picks

Type

create-skill

1.3kstars132forks

View on GitHub

How to install create-skill?

neolabhq/context-engineering-kit/create-skill

$npx -y skills add NeoLabHQ/context-engineering-kit --skill create-skill

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Create Skill Command
2
3	This command provides guidance for creating effective skills.
4
5	## Overview
6
7	Writing skills IS Test-Driven Development applied to process documentation.
8
9	Personal skills live in agent-specific directories (`~/.claude/skills` for Claude Code, `~/.codex/skills` for Codex)
10
11	You write test cases (pressure scenarios with subagents), watch them fail (baseline behavior), write the skill (documentation), watch tests pass (agents comply), and refactor (close loopholes).
12
13	Core principle: If you didn't watch an agent fail without the skill, you don't know if the skill teaches the right thing.
14
15	REQUIRED BACKGROUND: You MUST understand Test-Driven Development before using this skill. That skill defines the fundamental RED-GREEN-REFACTOR cycle. This skill adapts TDD to documentation.
16
17	Official guidance: The Anthropic's official skill authoring best practices provided at the `/apply-anthropic-skill-best-practices` command, they enhance `prompt-engineering` skill. Use skill and the document, as they not copy but add to each other. These document provides additional patterns and guidelines that complement the TDD-focused approach in this skill.
18
19	## About Skills
20
21	Skills are modular, self-contained packages that extend Claude's capabilities by providing
22	specialized knowledge, workflows, and tools. Think of them as "onboarding guides" for specific
23	domains or tasks—they transform Claude from a general-purpose agent into a specialized agent
24	equipped with procedural knowledge that no model can fully possess.
25
26	## What is a Skill?
27
28	A skill is a reference guide for proven techniques, patterns, or tools. Skills help future Claude instances find and apply effective approaches.
29
30	Skills are: Reusable techniques, patterns, tools, reference guides
31
32	Skills are NOT: Narratives about how you solved a problem once
33
34	### What Skills Provide
35
36	1. Specialized workflows - Multi-step procedures for specific domains
37	2. Tool integrations - Instructions for working with specific file formats or APIs
38	3. Domain expertise - Company-specific knowledge, schemas, business logic
39	4. Bundled resources - Scripts, references, and assets for complex and repetitive tasks
40
41	## TDD Mapping for Skills
42
43	\| TDD Concept \| Skill Creation \|
44	\|-------------\|----------------\|
45	\| Test case \| Pressure scenario with subagent \|
46	\| Production code \| Skill document (SKILL.md) \|
47	\| Test fails (RED) \| Agent violates rule without skill (baseline) \|
48	\| Test passes (GREEN) \| Agent complies with skill present \|
49	\| Refactor \| Close loopholes while maintaining compliance \|
50	\| Write test first \| Run baseline scenario BEFORE writing skill \|
51	\| Watch it fail \| Document exact rationalizations agent uses \|
52	\| Minimal code \| Write skill addressing those specific violations \|
53	\| Watch it pass \| Verify agent now complies \|
54	\| Refactor cycle \| Find new rationalizations → plug → re-verify \|
55
56	The entire skill creation process follows RED-GREEN-REFACTOR.
57
58	## When to Create a Skill
59
60	Create when:
61
62	- Technique wasn't intuitively obvious to you
63	- You'd reference this again across projects
64	- Pattern applies broadly (not project-specific)
65	- Others would benefit
66
67	Don't create for:
68
69	- One-off solutions
70	- Standard practices well-documented elsewhere
71	- Project-specific conventions (put in CLAUDE.md)
72
73	## Skill Types
74
75	### Technique
76
77	Concrete method with steps to follow (condition-based-waiting, root-cause-tracing)
78
79	### Pattern
80
81	Way of thinking about problems (flatten-with-flags, test-invariants)
82
83	### Reference
84
85	API docs, syntax guides, tool documentation (office docs)
86
87	## Directory Structure
88
89	```
90	skills/
91	skill-name/
92	SKILL.md # Main reference (required)
93	supporting-file.* # Only if needed
94	```
95
96	Flat namespace - all skills in one searchable namespace
97
98	Separate files for:
99
100	1. Heavy reference (100+ lines) - API docs, comprehensive syntax
101	2. Reusable tools - Scripts, utilities, templates
102
103	Keep inline:
104
105	- Principles and concepts
106	- Code patterns (< 50 lines)
107	- Everything else
108
109	## Anatomy of a Skill
110
111	Every skill consists of a required SKILL.md file and optional bundled resources:
112
113	```
114	skill-name/
115	├── SKILL.md (required)
116	│ ├── YAML frontmatter metadata (required)
117	│ │ ├── name: (required)
118	│ │ └── description: (required)
119	│ └── Markdown instructions (required)
120	└── Bundled Resources (optional)
121	├── scripts/ - Executable code (Python/Bash/etc.)
122	├── references/ - Docu

People who liked this also viewed

6 picks

Type