Repository inventory

davidosemwegie/doxy

Skills indexed from this repository, with install-style signals scoped to the repo.
1 skills4 GitHub stars0 weekly installsTypeScriptGitHubOwner profile

Overview

This skill converts any documentation website into ready-to-use agent skills for Claude Code. It crawls a docs site, extracts navigation and page content, and generates a structured skill folder with per-page skill documents and a manifest. The output is optimized for practical, actionable guidance and quick agent integration.

How this skill works

The skill fetches the documentation URL, extracts the library name and the full sidebar navigation, then walks each navigation item to pull focused technical content. For each page it summarizes the title, produces a concise summary, and captures key concepts, APIs, and examples as structured skill documents. A manifest records source metadata and processing details; errors and problematic pages are reported.

When to use it

  • You need to turn a documentation site into agent skills for automation or code assistance.
  • You want an agent to answer questions about a library using the library’s official docs.
  • You need an indexed, navigable set of task-focused skill documents derived from docs.
  • You want to bulk-generate actionable guidance and code examples from a docs site.
  • You need to onboard an AI assistant on a proprietary or open-source API quickly.

Best practices

  • Provide the root documentation URL that contains a sidebar navigation; relative links are required.
  • Validate that the docs use a stable sidebar structure before running a full crawl.
  • Run in batches for very large docs and review generated skills for quality and completeness.
  • Keep each generated skill concise (under ~200 lines); move extended content to a references subfolder.
  • Use imperative, action-focused language and include code examples when available.

Example use cases

  • Create a skills set for an internal API docs site to enable an assistant to answer developer questions.
  • Convert an open-source library’s documentation into agent-ready skills to accelerate onboarding.
  • Generate targeted skills for key topics (installation, API reference, tutorials) to embed into a help chatbot.
  • Crawl product docs and produce a manifest and per-topic skill files for automated testing or QA assistance.

FAQ

Failed pages are logged and processing continues; the report lists pages that failed so you can re-run or inspect them manually.

Can large pages be included?

Large pages are split: a concise skill document is created and full details move into a references subfolder to keep skills readable and within size limits.

How is a library name chosen?

The library name is extracted from the docs and normalized to lowercase kebab-case to form the skill folder and manifest name.

1 skills

More from this maintainer
Other repositories and skills published under the same GitHub owner.
Skills library
Jump back to the full directory or explore grouped topics.
Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational