Stats
Actions
Tags
From omni-sw
Agentic web scraper that reads API documentation from a URL, extracts endpoints, authentication, models, and error handling into a structured corpus, then invokes /skill-creator to generate a new skill that serves as an interactive reference for that API. Use when: (1) user provides a URL to an API documentation site, (2) user says "scrape api docs", "create api reference skill", "import api documentation", "api doc scraper", (3) user wants to create a skill from external API documentation for future technical Q&A.
How this skill is triggered — by the user, by Claude, or both
Slash command
/omni-sw:api-doc-scraperThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Scrape web-based API documentation and generate a Claude Code skill that serves as an interactive technical reference for the scraped API.
Scrape web-based API documentation and generate a Claude Code skill that serves as an interactive technical reference for the scraped API.
<url> (required) — Root URL of the API documentation--name <skill-name> (optional) — Override the generated skill name (default: derived from API name)| File | Description | Load when |
|---|---|---|
references/workflow-phases.md | Detailed instructions for all 4 phases (Discovery, Scraping, Skill Generation, Optimization) | Executing any phase |
references/scraping-structure.md | Corpus format template and content extraction heuristics | Building or validating the corpus structure |
The scraper operates in 4 phases:
nav-map.jsonscripts/scrape_api_docs.py to produce corpus.md (outside context window)corpus.md, invoke /skill-creator with the content/apply-progressive-disclosure optimize on the outputFetch the root URL, extract navigation links via WebFetch, and write nav-map.json. See references/workflow-phases.md → Phase 1 for the full WebFetch prompt and steps.
Run scripts/scrape_api_docs.py to fetch all pages and produce corpus.md. See references/workflow-phases.md → Phase 2 for the full command, script behavior, and error recovery.
Derive the skill name, read the corpus, and invoke /skill-creator with the structured prompt. See references/workflow-phases.md → Phase 3 for the full template and instructions.
Run /apply-progressive-disclosure optimize on the generated skill and clean up temporary files. See references/workflow-phases.md → Phase 4 for details.
When complete, report to the user:
/{skill-name} "How do I authenticate?" or similarnpx claudepluginhub vitorandtxr/omni-sw --plugin omni-swIngests documentation site URLs, discovers pages via sitemap or nav crawl, extracts markdown, and generates Claude Code skill packages with SKILL.md indexes and references.
Converts documentation websites, GitHub repos, and PDFs into Claude AI skills automatically.
Automatically detects source types (docs, repos, PDFs, videos) and builds AI skills using Skill Seekers MCP tools. Use for converting knowledge sources into skill packages.