skill_seekers

Convert documentation sites, repositories, PDFs, videos, and codebases into AI skills.

Updated May 3, 2026
One-click install
npx skills add https://github.com/miethe/boxbrain-2 --skill skill-seekers-miethe
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill_seekers
Source: https://github.com/miethe/boxbrain-2/tree/main/.claude/skills/skill_seekers
Command: npx skills add https://github.com/miethe/boxbrain-2 --skill skill-seekers-miethe

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Skill Seekers reduces the effort and inconsistency involved in converting documentation, repositories, PDFs, videos, and codebases into usable AI skills.

Core Features & Use Cases

  • Source Detection: Identifies source types and selects the appropriate MCP tool or CLI workflow.
  • Skill Generation: Scrapes, enhances, reorganizes, packages, and installs progressively disclosed skills.
  • Quality Assurance: Verifies page coverage, detects contaminated scrape output, scores quality, and enforces metadata consistency.
  • Use Case: Rebuild a Claude Code skill from a documentation site by sourcing its llms-full.txt file, splitting content into topic references, validating the results, and packaging the finished skill.

Quick Start

Use the skill_seekers skill to create a high-quality Claude Code skill from the provided documentation site or repository, verify its output, and package it for installation.

Frequently Asked Questions about skill_seekers

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a Claude Code skill from a documentation site?

You can build a Claude Code skill by detecting the documentation source, prioritizing its llms.txt file, splitting content into topic references, validating page coverage, and packaging the finished skill for installation.

What is progressive disclosure when generating AI skills from repositories?

Progressive disclosure is a skill generation approach that splits scraped documentation or repository content into topic references, ensuring the AI retrieves information in layers without overwhelming the context window.

Can I use skill generation tools to convert local codebases and PDFs into AI skills?

Yes, skill generation supports converting local codebases, PDFs, and videos by detecting the source type and selecting the appropriate MCP tool or CLI workflow to scrape, enhance, and package the content.

How do you validate the quality of scraped documentation for AI skills?

Quality validation checks sitemap coverage, detects contaminated scrape output, scores overall quality, and enforces metadata consistency to ensure the generated skill is reliable.

What is the best way to handle fallback when scraping documentation sites for skills?

When scraping fails, the best fallback approach utilizes MCP or CLI fallback handling mechanisms after initial source-type detection, ensuring continuous content extraction for skill generation.

Does skill generation support exporting packaged skills to a vector database?

Yes, skill generation supports vector database export, allowing progressively disclosed skills rebuilt from GitHub repositories or documentation to be directly exported for retrieval.