claude-skills-cli 0.0.4 → 0.0.5

package/docs/SKILL-DEVELOPMENT.md

@@ -15,7 +15,8 @@ claude-skills package .claude/skills/my-skill
 
 ## The 6-Step Process
 
-Based on Anthropic's skill-creator methodology, adapted for devhub-crm.
+Based on Anthropic's skill-creator methodology, adapted for
+devhub-crm.
 
 ### Step 1: Understanding with Concrete Examples
 
@@ -92,7 +93,8 @@ claude-skills init \
 
 ### Step 4: Editing the Skill
 
-**Focus**: Write content for another instance of Claude to use effectively.
+**Focus**: Write content for another instance of Claude to use
+effectively.
 
 #### Start with Reusable Contents
 
@@ -315,7 +317,8 @@ Add comments explaining WHY, not just WHAT:
 ```markdown
 ## ID Generation
 
-Generate IDs with nanoid() to ensure uniqueness without database overhead.
+Generate IDs with nanoid() to ensure uniqueness without database
+overhead.
 
 <!-- Not just: "Use nanoid()" -->
 ```
@@ -328,7 +331,8 @@ Always run validation:
 claude-skills validate .claude/skills/my-skill --strict
 ```
 
-Strict mode treats warnings as errors - use before packaging for distribution.
+Strict mode treats warnings as errors - use before packaging for
+distribution.
 
 ---
 
@@ -439,7 +443,7 @@ description: Brief description including when to use this skill
  */
 
 function main() {
-  // Script logic here
+	// Script logic here
 }
 
 main();
@@ -449,7 +453,8 @@ main();
 
 ## Next Steps
 
-1. Read [SKILLS-ARCHITECTURE.md](SKILLS-ARCHITECTURE.md) for system overview
+1. Read [SKILLS-ARCHITECTURE.md](SKILLS-ARCHITECTURE.md) for system
+   overview
 2. See [SKILL-EXAMPLES.md](SKILL-EXAMPLES.md) for real examples
 3. Create your first skill with `claude-skills init`
 4. Join skill development workflow for devhub-crm

package/docs/SKILL-EXAMPLES.md

@@ -1,6 +1,7 @@
 # Skill Examples
 
-Real-world examples from Anthropic's Skills repository and Claude Cookbooks, with analysis of what makes them effective.
+Real-world examples from Anthropic's Skills repository and Claude
+Cookbooks, with analysis of what makes them effective.
 
 ---
 
@@ -21,15 +22,19 @@ applying-brand-guidelines/
 ```yaml
 ---
 name: applying-brand-guidelines
-description: This skill applies consistent corporate branding and styling to all generated documents including colors, fonts, layouts, and messaging
+description:
+  This skill applies consistent corporate branding and styling to all
+  generated documents including colors, fonts, layouts, and messaging
 ---
 ```
 
 ### What Makes It Good
 
-**✅ Clear Scope**: Focuses on one thing - brand consistency across documents
+**✅ Clear Scope**: Focuses on one thing - brand consistency across
+documents
 
-**✅ Actionable Content**: Specific hex codes, font sizes, spacing rules
+**✅ Actionable Content**: Specific hex codes, font sizes, spacing
+rules
 
 ```markdown
 ### Color Palette
@@ -45,7 +50,8 @@ description: This skill applies consistent corporate branding and styling to all
 // Checks documents for brand compliance automatically
 ```
 
-**✅ "When to Use" in Description**: "applies consistent corporate branding...to all generated documents"
+**✅ "When to Use" in Description**: "applies consistent corporate
+branding...to all generated documents"
 
 ### Key Takeaway
 
@@ -76,7 +82,10 @@ pdf/
 ````markdown
 ---
 name: pdf
-description: Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files or when the user mentions PDFs, forms, or document extraction.
+description:
+  Extract text and tables from PDF files, fill forms, merge documents.
+  Use when working with PDF files or when the user mentions PDFs,
+  forms, or document extraction.
 ---
 
 # PDF Processing
@@ -91,7 +100,7 @@ const pdf = require('pdf-parse');
 
 const dataBuffer = fs.readFileSync('document.pdf');
 pdf(dataBuffer).then((data) => {
-  console.log(data.text);
+	console.log(data.text);
 });
 ```
 ````
@@ -126,14 +135,9 @@ Large domains benefit from splitting:
 ### Structure (from Claude Cookbooks)
 ```
 
-financial-analyzer/
-├── SKILL.md
-├── references/
-│ ├── formulas.md
-│ └── ratios-reference.md
-└── scripts/
-├── calculate_ratios.js
-└── generate_dashboard.js
+financial-analyzer/ ├── SKILL.md ├── references/ │ ├── formulas.md │
+└── ratios-reference.md └── scripts/ ├── calculate_ratios.js └──
+generate_dashboard.js
 
 ````
 
@@ -161,7 +165,8 @@ const currentRatio = currentAssets / currentLiabilities;
 const quickRatio = (currentAssets - inventory) / currentLiabilities;
 ```
 
-For complete ratio formulas, see [references/formulas.md](references/formulas.md).
+For complete ratio formulas, see
+[references/formulas.md](references/formulas.md).
 
 ````
 
@@ -208,7 +213,10 @@ database-schema/
 ````markdown
 ---
 name: database-schema
-description: Complete database schema with table structures, relationships, and indexes for the project. Use when writing SQL queries, understanding data models, or working with database operations.
+description:
+  Complete database schema with table structures, relationships, and
+  indexes for the project. Use when writing SQL queries, understanding
+  data models, or working with database operations.
 ---
 
 # Database Schema
@@ -263,16 +271,9 @@ Reference-heavy skills should:
 ### Structure (from Anthropic examples)
 ```
 
-component-library/
-├── SKILL.md
-├── references/
-│ ├── button-variants.md
-│ ├── form-patterns.md
-│ └── layout-components.md
-└── assets/
-└── component-templates/
-├── button.tsx
-└── form.tsx
+component-library/ ├── SKILL.md ├── references/ │ ├──
+button-variants.md │ ├── form-patterns.md │ └── layout-components.md
+└── assets/ └── component-templates/ ├── button.tsx └── form.tsx
 
 ````
 
@@ -331,15 +332,9 @@ Component library skills should:
 ### Structure
 ```
 
-github-api/
-├── SKILL.md
-├── references/
-│ ├── endpoints.md
-│ ├── authentication.md
-│ └── rate-limits.md
-└── scripts/
-├── test_connection.js
-└── check_rate_limit.js
+github-api/ ├── SKILL.md ├── references/ │ ├── endpoints.md │ ├──
+authentication.md │ └── rate-limits.md └── scripts/ ├──
+test_connection.js └── check_rate_limit.js
 
 ````
 
@@ -371,7 +366,8 @@ Check rate limits before making requests:
 node scripts/check_rate_limit.js
 ```
 
-For complete API reference, see [references/endpoints.md](references/endpoints.md).
+For complete API reference, see
+[references/endpoints.md](references/endpoints.md).
 
 ````
 
@@ -410,7 +406,10 @@ description: Helps with database operations
 
 ```yaml
 ---
-description: SQLite query patterns using better-sqlite3 for contacts, companies, and interactions tables. Use when writing SELECT, INSERT, UPDATE, or DELETE operations with prepared statements.
+description:
+  SQLite query patterns using better-sqlite3 for contacts, companies,
+  and interactions tables. Use when writing SELECT, INSERT, UPDATE, or
+  DELETE operations with prepared statements.
 ---
 ```
 
@@ -437,8 +436,8 @@ description: SQLite query patterns using better-sqlite3 for contacts, companies,
 
 [10 common patterns]
 
-For complete schema: [references/schema.md](references/schema.md)
-For all examples: [references/queries.md](references/queries.md)
+For complete schema: [references/schema.md](references/schema.md) For
+all examples: [references/queries.md](references/queries.md)
 ```
 
 ### ❌ Anti-Pattern 3: Using Second Person
@@ -470,7 +469,10 @@ description: Handle form submissions
 
 ```yaml
 ---
-description: Handle form submissions with validation, error handling, and reactive updates. Use when implementing forms, processing user input, or validating data before database operations.
+description:
+  Handle form submissions with validation, error handling, and
+  reactive updates. Use when implementing forms, processing user
+  input, or validating data before database operations.
 ---
 ```
 
@@ -480,7 +482,8 @@ description: Handle form submissions with validation, error handling, and reacti
 
 ### Example: Multi-Skill Workflow
 
-**User Request**: "Create a GitHub contact dashboard with database stats"
+**User Request**: "Create a GitHub contact dashboard with database
+stats"
 
 **Skills Activated**:
 
@@ -489,7 +492,8 @@ description: Handle form submissions with validation, error handling, and reacti
 3. `sveltekit-patterns` - Build component structure
 4. `daisyui-conventions` - Apply styling
 
-**Why This Works**: Each skill handles its domain, Claude composes them naturally.
+**Why This Works**: Each skill handles its domain, Claude composes
+them naturally.
 
 ---
 
@@ -498,8 +502,10 @@ description: Handle form submissions with validation, error handling, and reacti
 ### From Anthropic Skills
 
 1. **Metadata drives discovery** - Spend time on descriptions
-2. **Progressive disclosure saves tokens** - Don't front-load everything
-3. **Scripts for determinism** - Code that doesn't change shouldn't be generated
+2. **Progressive disclosure saves tokens** - Don't front-load
+   everything
+3. **Scripts for determinism** - Code that doesn't change shouldn't be
+   generated
 4. **References for depth** - Keep SKILL.md navigable
 5. **Real examples** - Pull from actual codebases
 

package/docs/SKILLS-ARCHITECTURE.md

@@ -2,7 +2,10 @@
 
 ## Overview
 
-Claude Skills are modular capabilities that extend Claude's functionality through a filesystem-based architecture. They provide specialized domain expertise, workflows, and tools that transform Claude from a general-purpose agent into a specialist.
+Claude Skills are modular capabilities that extend Claude's
+functionality through a filesystem-based architecture. They provide
+specialized domain expertise, workflows, and tools that transform
+Claude from a general-purpose agent into a specialist.
 
 ## Core Concepts
 
@@ -52,8 +55,8 @@ Claude sees this at startup and knows:
 - When to use it
 - How to trigger it
 
-**Token cost**: ~100 tokens per skill (metadata only)
-**When**: At agent startup, every conversation
+**Token cost**: ~100 tokens per skill (metadata only) **When**: At
+agent startup, every conversation
 
 ### Level 2: Instructions (<5k tokens)
 
@@ -71,21 +74,21 @@ The markdown body of SKILL.md contains:
 ## Core Principles
 
 - Use prepared statements for all queries
-- Generate IDs with nanoid()
-  ...
+- Generate IDs with nanoid() ...
 
 For complete schema, see [references/schema.md](references/schema.md)
 ```
 
-**Token cost**: ~3-5k tokens typically
-**When**: Only when Claude determines skill is relevant
+**Token cost**: ~3-5k tokens typically **When**: Only when Claude
+determines skill is relevant
 
 ### Level 3: Resources (unlimited)
 
 **Loaded as needed**
 
 - **references/**: Documentation Claude reads into context as needed
-- **scripts/**: Executable code Claude runs without loading into context
+- **scripts/**: Executable code Claude runs without loading into
+  context
 - **assets/**: Files used in output (templates, images, fonts)
 
 ```bash
@@ -226,7 +229,8 @@ Detailed documentation loaded only when needed by Claude.
 ```markdown
 # In SKILL.md
 
-For complete database schema with all relationships, see [references/schema.md](references/schema.md).
+For complete database schema with all relationships, see
+[references/schema.md](references/schema.md).
 
 # Claude can then:
 
@@ -237,7 +241,8 @@ cat references/schema.md # Load when needed
 
 ### Purpose
 
-Executable code for deterministic operations that don't need token generation.
+Executable code for deterministic operations that don't need token
+generation.
 
 ### When to Use Scripts
 
@@ -309,7 +314,8 @@ Schema reference (if needed): ~2000 tokens
 Total: 100 tokens normally, 3100-5100 when used
 ```
 
-**Savings**: Skill metadata is loaded once. Without skill, you pay ~3500 tokens every conversation even if not needed.
+**Savings**: Skill metadata is loaded once. Without skill, you pay
+~3500 tokens every conversation even if not needed.
 
 ## Skill Composition
 
@@ -354,27 +360,22 @@ Each skill loads independently, shares context naturally.
 
 ### Do:
 
-✅ Keep SKILL.md concise and actionable
-✅ Use imperative voice for instructions
-✅ Provide concrete examples
-✅ Link to references for details
-✅ Include "when to use" in description
-✅ Use scripts for deterministic operations
-✅ Group related content in references
-✅ Test skills on real tasks
+✅ Keep SKILL.md concise and actionable ✅ Use imperative voice for
+instructions ✅ Provide concrete examples ✅ Link to references for
+details ✅ Include "when to use" in description ✅ Use scripts for
+deterministic operations ✅ Group related content in references ✅
+Test skills on real tasks
 
 ### Don't:
 
-❌ Duplicate content between SKILL.md and references
-❌ Use second person ("you")
-❌ Include entire documentation inline
-❌ Forget to specify when to use skill
-❌ Make descriptions too generic
-❌ Leave TODO placeholders
-❌ Skip validation before packaging
+❌ Duplicate content between SKILL.md and references ❌ Use second
+person ("you") ❌ Include entire documentation inline ❌ Forget to
+specify when to use skill ❌ Make descriptions too generic ❌ Leave
+TODO placeholders ❌ Skip validation before packaging
 
 ## Next Steps
 
-- Read [SKILL-DEVELOPMENT.md](SKILL-DEVELOPMENT.md) for creation workflow
+- Read [SKILL-DEVELOPMENT.md](SKILL-DEVELOPMENT.md) for creation
+  workflow
 - See [SKILL-EXAMPLES.md](SKILL-EXAMPLES.md) for real-world examples
 - Use `claude-skills init` to create your first skill

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "claude-skills-cli",
-  "version": "0.0.4",
+  "version": "0.0.5",
   "description": "CLI toolkit for creating and managing Claude Agent Skills",
   "type": "module",
   "main": "./dist/index.js",
@@ -10,7 +10,6 @@
   "files": [
     "dist",
     "docs",
-    "skills",
     "README.md"
   ],
   "engines": {
@@ -47,6 +46,7 @@
   },
   "scripts": {
     "build": "tsc",
+    "postbuild": "mkdir -p dist/skills && cp -r .claude/skills/skill-creator dist/skills/skill-creator",
     "dev": "tsc --watch",
     "start": "node ./dist/index.js",
     "format": "prettier --write .",

package/skills/skill-creator/SKILL.md

@@ -1,106 +0,0 @@
----
-name: skill-creator
-description: Create Claude Skills using claude-skills-cli. Use when building new skills, running validation, or packaging skills for distribution with TypeScript/Node.
----
-
-# Skill Creator
-
-Create effective Claude Skills using the `claude-skills-cli` tool (TypeScript/Node).
-
-## Quick Start
-
-```bash
-# Create skill
-npx claude-skills init --name my-skill \
-  --description "What it does. Use when [trigger keywords]"
-
-# Validate
-npx claude-skills validate .claude/skills/my-skill
-
-# Package
-npx claude-skills package .claude/skills/my-skill
-```
-
-## Progressive Disclosure System
-
-Skills load in 3 levels:
-
-| Level                       | When Loaded    | Token Budget |
-| --------------------------- | -------------- | ------------ |
-| **Level 1** - Metadata      | Always         | ~100 tokens  |
-| **Level 2** - SKILL.md body | When triggered | <5k tokens   |
-| **Level 3** - Bundled files | As needed      | Unlimited    |
-
-**Key principle**: Keep Level 1 & 2 lean. Move details to Level 3.
-
-## CLI Commands
-
-### init - Create Skill
-
-```bash
-npx claude-skills init --name skill-name \
-  --description "Brief description with trigger keywords"
-```
-
-Creates: `SKILL.md`, `README.md`, `references/`
-
-### validate - Check Quality
-
-```bash
-npx claude-skills validate .claude/skills/skill-name
-npx claude-skills validate .claude/skills/skill-name --strict
-```
-
-Checks progressive disclosure compliance.
-
-### package - Create Zip
-
-```bash
-npx claude-skills package .claude/skills/skill-name
-```
-
-Creates uploadable zip for Claude.ai or API.
-
-## Writing Tips
-
-**Level 1 (Description)**:
-
-- Format: `[Tech] + [Operations] + [Trigger]`
-- Target: <200 chars
-- Include: "Use when...", "Use for...", "Use to..."
-
-**Level 2 (SKILL.md Body)**:
-
-- Target: ~50 lines
-- Structure: Quick Start, Core Patterns (3-5), Links to references
-- Voice: Imperative ("Use X"), not second person ("You should use X")
-
-**Level 3 (References)**:
-
-- Move detailed docs to `references/`
-- Link from SKILL.md
-- Unlimited size
-
-## Reference Documentation
-
-For detailed guidance:
-
-- [cli-reference.md](references/cli-reference.md) - Complete CLI commands and options
-- [cli-feedback.md](references/cli-feedback.md) - Real-world usage patterns and tips
-- [anthropic-resources.md](references/anthropic-resources.md) - Official Anthropic best practices
-- [writing-guide.md](references/writing-guide.md) - Voice, structure, and code examples
-- [skill-examples.md](references/skill-examples.md) - Example skills and patterns
-
-## Notes
-
-- Skills are iterative - start minimal, refine based on usage
-- Description drives discovery - make it keyword-rich
-- Validate frequently during development
-- Test in real conversations
-
-<!--
-PROGRESSIVE DISCLOSURE:
-- This is Level 2 - keep lean and scannable
-- Move detailed content to references/
-- Target: ~50 lines for optimal scannability
--->