Bootstrap Skill - Self-Hosting (v3.6.0)
Version: 3.6.0 Feature: Bootstrap Skill (Dogfooding) Status: ✅ Production Ready Last Updated: 2026-02-18
Overview
The Bootstrap Skill feature allows Skill Seekers to analyze itself and generate a Claude Code skill containing its own documentation, API reference, code patterns, and usage examples. This is the ultimate form of "dogfooding" - using the tool to document itself.
What You Get:
- Complete Skill Seekers documentation as a Claude Code skill
- CLI command reference with examples
- Auto-generated API documentation from codebase
- Design pattern detection from source code
- Test example extraction for learning
- Installation into Claude Code for instant access
Use Cases:
- Learn Skill Seekers by having it explain itself to Claude
- Quick reference for CLI commands while working
- API documentation for programmatic usage
- Code pattern examples from the source
- Self-documenting development workflow
Quick Start
One-Command Installation
# Generate and install the bootstrap skill
./scripts/bootstrap_skill.shThis script will:
- ✅ Analyze the Skill Seekers codebase (C3.x features)
- ✅ Merge handcrafted header with auto-generated content
- ✅ Validate YAML frontmatter and structure
- ✅ Create
output/skill-seekers/directory - ✅ Install to Claude Code (optional)
Time: ~2-5 minutes (depending on analysis depth)
Manual Installation
# 1. Run codebase analysis
skill-seekers codebase \
--directory . \
--output output/skill-seekers \
--name skill-seekers
# 2. Merge with custom header (optional)
cat scripts/skill_header.md output/skill-seekers/SKILL.md > output/skill-seekers/SKILL_MERGED.md
mv output/skill-seekers/SKILL_MERGED.md output/skill-seekers/SKILL.md
# 3. Install to Claude Code
skill-seekers install-agent \
--skill-dir output/skill-seekers \
--agent-dir ~/.claude/skills/skill-seekersHow It Works
Architecture
The bootstrap skill combines three components:
┌─────────────────────────────────────────────────────────┐
│ Bootstrap Skill Architecture │
├─────────────────────────────────────────────────────────┤
│ │
│ 1. Handcrafted Header (scripts/skill_header.md) │
│ ├── YAML frontmatter │
│ ├── Installation instructions │
│ ├── Quick start guide │
│ └── Core concepts │
│ │
│ 2. Auto-Generated Content (codebase_scraper.py) │
│ ├── C3.1: Design pattern detection │
│ ├── C3.2: Test example extraction │
│ ├── C3.3: How-to guide generation │
│ ├── C3.4: Configuration extraction │
│ ├── C3.5: Architectural overview │
│ ├── C3.7: Architectural pattern detection │
│ ├── C3.8: API reference + dependency graphs │
│ └── Code analysis (9 languages) │
│ │
│ 3. Validation System (frontmatter detection) │
│ ├── YAML frontmatter check │
│ ├── Required field validation │
│ └── Structure verification │
│ │
└─────────────────────────────────────────────────────────┘Step 1: Codebase Analysis
The codebase_scraper.py module analyzes the Skill Seekers source code:
skill-seekers codebase --directory . --output output/skill-seekersWhat Gets Analyzed:
- Python source files (
src/skill_seekers/**/*.py) - Test files (
tests/**/*.py) - Configuration files (
configs/*.json) - Documentation (
docs/**/*.md,README.md, etc.)
C3.x Features Applied:
- C3.1: Detects design patterns (Strategy, Factory, Singleton, etc.)
- C3.2: Extracts test examples showing real usage
- C3.3: Generates how-to guides from test workflows
- C3.4: Extracts configuration patterns (CLI args, env vars)
- C3.5: Creates architectural overview of the codebase
- C3.7: Detects architectural patterns (MVC, Repository, etc.)
- C3.8: Builds API reference and dependency graphs
Step 2: Header Combination
The bootstrap script merges a handcrafted header with auto-generated content:
# scripts/bootstrap_skill.sh does this:
cat scripts/skill_header.md output/skill-seekers/SKILL.md > merged.mdWhy Two Parts?
- Header: Curated introduction, installation steps, core concepts
- Auto-generated: Always up-to-date code patterns, examples, API docs
Header Structure (scripts/skill_header.md):
---
name: skill-seekers
version: 3.6.0
description: |
Documentation-to-AI skill conversion tool. Use when working with
Skill Seekers codebase, CLI commands, or API integration.
tags: [documentation, scraping, ai-skills, mcp]
---
# Skill Seekers - Documentation to AI Skills
## Installation
...
## Quick Start
...
## Core Concepts
...
<!-- AUTO-GENERATED CONTENT STARTS HERE -->Step 3: Validation
The bootstrap script validates the final skill:
# Check for YAML frontmatter
if ! grep -q "^---$" output/skill-seekers/SKILL.md; then
echo "❌ Missing YAML frontmatter"
exit 1
fi
# Validate required fields
python -c "
import yaml
with open('output/skill-seekers/SKILL.md') as f:
content = f.read()
frontmatter = yaml.safe_load(content.split('---')[1])
required = ['name', 'version', 'description']
for field in required:
assert field in frontmatter, f'Missing {field}'
"Validated Fields:
- ✅
name- Skill name - ✅
version- Version number - ✅
description- When to use this skill - ✅
tags- Categorization tags - ✅ Proper YAML syntax
- ✅ Content structure
Step 4: Output
The final skill is created in output/skill-seekers/:
output/skill-seekers/
├── SKILL.md # Main skill file (300-500 lines)
├── references/ # Detailed references
│ ├── api_reference/ # API documentation
│ │ ├── doc_scraper.md
│ │ ├── github_scraper.md
│ │ └── ...
│ ├── patterns/ # Design patterns detected
│ │ ├── strategy_pattern.md
│ │ ├── factory_pattern.md
│ │ └── ...
│ ├── test_examples/ # Usage examples from tests
│ │ ├── scraping_examples.md
│ │ ├── packaging_examples.md
│ │ └── ...
│ └── how_to_guides/ # Generated guides
│ ├── how_to_scrape_docs.md
│ ├── how_to_package_skills.md
│ └── ...
└── metadata.json # Skill metadataAdvanced Usage
Customizing the Header
Edit scripts/skill_header.md to customize the introduction:
---
name: skill-seekers
version: 3.6.0
description: |
YOUR CUSTOM DESCRIPTION HERE
tags: [your, custom, tags]
custom_field: your_value
---
# Your Custom Title
Your custom introduction...
<!-- AUTO-GENERATED CONTENT STARTS HERE -->Guidelines:
- Keep frontmatter in YAML format
- Include required fields:
name,version,description - Add custom fields as needed
- Marker comment preserves auto-generated content location
Validation Options
The bootstrap script supports custom validation rules:
# scripts/bootstrap_skill.sh (excerpt)
# Custom validation function
validate_skill() {
local skill_file=$1
# Check frontmatter
if ! has_frontmatter "$skill_file"; then
echo "❌ Missing frontmatter"
return 1
fi
# Check required fields
if ! has_required_fields "$skill_file"; then
echo "❌ Missing required fields"
return 1
fi
# Check content structure
if ! has_proper_structure "$skill_fi