Configuration

There are three ways to configure storybook-to-skills-md: configuration files, environment variables, and CLI flags. They are prioritized in that order (CLI flags override everything).

Configuration File

Create a .skillgenrc.json file in your project root:

{
  "storybookUrl": "https://your-storybook.com",
  "indexFile": "./storybook-static/index.json",
  "sourceDir": "./src/components",
  "outputDir": "./skills",
  "provider": "openai",
  "model": "gpt-4o",
  "apiKey": "sk-...",
  "concurrency": 3,
  "timeout": 60000,
  "retries": 2,
  "fetchRetries": 3,
  "extractionConcurrency": 3,
  "include": ["Components/**", "Patterns/**"],
  "exclude": ["**/Internal/**", "**/Deprecated/**"],
  "verbose": false,
  "dryRun": false,
  "force": false,
  "serverOnly": false,
  "indexSkill": true,
  "indexSkillTemplate": "./templates/index-skill.md",
  "promptFile": "./prompts/custom.md",
  "logPrompts": "./debug-logs"
}

Custom Config Path

You can specify a custom config file path:

storybook-to-skills-md generate --config ./config/custom.json

Environment Variables

All configuration options can be set via environment variables with the SKILLGEN_ prefix:

export SKILLGEN_STORYBOOK_URL=https://your-storybook.com
export SKILLGEN_INDEX_FILE=./storybook-static/index.json
export SKILLGEN_SOURCE_DIR=./src/components
export SKILLGEN_OUTPUT_DIR=./skills
export SKILLGEN_PROVIDER=openai
export SKILLGEN_MODEL=gpt-4o
export SKILLGEN_API_KEY=sk-...
export SKILLGEN_CONCURRENCY=3
export SKILLGEN_TIMEOUT=60000
export SKILLGEN_RETRIES=2
export SKILLGEN_FETCH_RETRIES=3
export SKILLGEN_EXTRACTION_CONCURRENCY=3
export SKILLGEN_VERBOSE=true
export SKILLGEN_PROMPT_FILE=./prompts/custom.md

Note: Array options like include and exclude are not supported via environment variables. Use a config file for these.

CLI Flags

CLI flags override both config files and environment variables. See the Commands reference for all available flags.

Priority Order

Configuration is resolved in this order (later overrides earlier):

1. Default values (hardcoded in the tool)
2. Configuration file (.skillgenrc.json or --config)
3. Environment variables (SKILLGEN_*)
4. CLI flags (highest priority)

Example: Mixed Configuration

You can mix and match approaches. For example:

.skillgenrc.json:

{
  "sourceDir": "./src/components",
  "outputDir": "./skills",
  "provider": "openai",
  "model": "gpt-4o",
  "concurrency": 3,
  "include": ["Components/**"],
  "exclude": ["**/Internal/**"]
}

Environment:

export SKILLGEN_API_KEY=sk-...

Command:

storybook-to-skills-md generate --storybook-url https://storybook.example.com

This keeps sensitive data (API keys) in environment variables, common settings in the config file, and variable options (like URLs) in CLI flags.

Best Practices

1. Store API keys in environment variables - Never commit them to version control
2. Use config files for project-specific settings - Commit .skillgenrc.json to your repo
3. Use CLI flags for one-off overrides - Great for testing different models or providers
4. Use .gitignore - Add .skillgenrc.json if it contains sensitive data

Common Configurations

CI/CD Environment

{
  "indexFile": "./storybook-static/index.json",
  "sourceDir": "./src",
  "outputDir": "./skills",
  "provider": "anthropic",
  "model": "claude-3-5-sonnet-20241022",
  "concurrency": 5,
  "force": true
}

Then use environment variable for API key:

export SKILLGEN_API_KEY=$ANTHROPIC_API_KEY

Local Development

{
  "storybookUrl": "http://localhost:6006",
  "sourceDir": "./src/components",
  "outputDir": "./skills",
  "provider": "openai",
  "model": "gpt-4o",
  "verbose": true,
  "logPrompts": "./debug"
}

Production

{
  "storybookUrl": "https://design-system.example.com",
  "sourceDir": "./packages/components/src",
  "outputDir": "./docs/skills",
  "provider": "anthropic",
  "model": "claude-3-5-sonnet-20241022",
  "concurrency": 5,
  "timeout": 120000,
  "retries": 3,
  "include": ["Components/**", "Patterns/**"],
  "exclude": ["**/Internal/**", "**/Experimental/**"]
}