claudeskills-fetcher

# ClaudeSkills Fetcher Fetches the complete skill catalog from ClaudeSkills Hub (claudeskills.info) and writes it to a local text file. ## What it does 1. Calls `GET /api/v1/meta` to discover total counts and type breakdown 2. For each type (skill, subagent, plugin, command, hook, memory-tool, automation, claude-md-example), paginates through `GET /api/v1/search?type={type}&limit=100&offset={n}&sort=stars` 3. De-duplicates all results by `slug` 4. Extracts: `name`, `subtitle` (empty), `brief` (description), `category`, `url` (source.url), `stars` 5. Outputs one skill per line, tab-separated, to `claudeskills-skills.txt` ## Usage Run the helper script from the skill directory: ```bash node .trae/skills/claudeskills-fetcher/scripts/fetch.js ``` The script will: 1. Fetch meta information (total items, type counts) 2. Iterate through all 8 types, paginating with offset=0,100,200,... 3. De-duplicate by slug across all types 4. Write results to `claudeskills-skills.txt` in the current working directory ## Output format ``` name\tsubtitle\tbrief\tcategory\turl\tstars ``` Example line: ``` Code Reviewer\t\tExpert code review subagent.\t\thttps://github.com/ChrisRoyse/610ClaudeSubagents/blob/main/code-reviewer.md\t596 ``` ## Parameters - `outputFile` -- destination path (default: `./claudeskills-skills.txt`) - `delayMs` -- delay between API requests in ms (default: 1500) - `limit` -- items per page (default: 100, max 100)