# Skill Builder

> Builds AI skills from documentation, repos, PDFs, videos, and other sources using the Skill Seekers MCP server, with source detection, scraping, enhancement, packaging, and vector DB export.

- Skill: `oyi77/skill-builder` (Agent Skill)
- Install (CLI): `npx skillmds@latest add oyi77/skill-builder`
- Raw SKILL.md: https://api.skillmd.com/api/skills/oyi77/skill-builder/raw
- Safety review: PASS (external: skill-scanner PASS, skillspector CAUTION)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Docs & Writing, AI & ML, Agent Building, Knowledge Bases
- Tags: Documentation, Github, Mcp, Pdf, Scraping, Skill Builder, Skill Seekers, Video
- License: Apache-2.0
- Author: oyi77 (https://skillmd.com/u/oyi77)
- Updated: 2026-08-22
- Page: https://skillmd.com/skills/oyi77/skill-builder

---



# Skill Builder

You have access to the Skill Seekers MCP server which provides 40 tools for converting knowledge sources into AI-ready skills.

## Overview

Build new agent skills with proper structure, triggers, and verification.

## When NOT to Use

- You need to edit an existing skill (use the skill's own editing workflow)
- The source is a conversational transcript with no clear structure (use manual writing)
- You're looking for pre-built skills (use `skill_view` or search the 1ai-skills library)

## Anti-Rationalization Table

| Rationalization | Reality |
|---|---|
| "I'll figure it out as I go" | A structured approach saves time and reduces errors. Follow the workflow in this skill rather than improvising. |
| "I already know this topic" | Familiarity breeds shortcuts. Use the checklist to verify you haven't missed critical steps. |
| "This doesn't apply to my situation" | The patterns here generalize across contexts. Adapt, don't skip — the underlying principles hold. |
| "One more tool will fix it" | Adding complexity rarely solves process gaps. Master the core workflow first. |

## When to Use

**Trigger phrases:**
- "skill builder"
- "Wants to create an AI skill from a documentation site, GitHub repo, PDF, video,"
- "Needs to convert documentation into a format suitable for LLM consumption"
- "Wants to update or sync existing skills with their source documentation"


Use this skill when the user:
- Wants to create an AI skill from a documentation site, GitHub repo, PDF, video, or other source
- Needs to convert documentation into a format suitable for LLM consumption
- Wants to update or sync existing skills with their source documentation
- Needs to export skills to vector databases (Weaviate, Chroma, FAISS, Qdrant)
- Asks about scraping, converting, or packaging documentation for AI

## Source Type Detection

Automatically detect the source type from user input:

| Input Pattern | Source Type | Tool to Use |
|---------------|-------------|-------------|
| `https://...` (not GitHub/YouTube) | Documentation | `scrape_docs` |
| `owner/repo` or `github.com/...` | GitHub | `scrape_github` |
| `*.pdf` | PDF | `scrape_pdf` |
| YouTube/Vimeo URL or video file | Video | `scrape_video` |
| Local directory path | Codebase | `scrape_codebase` |
| `*.ipynb`, `*.html`, `*.yaml` (OpenAPI), `*.adoc`, `*.pptx`, `*.rss`, `*.1`-`.8` | Various | `scrape_generic` |
| JSON config file | Unified | Use config with `scrape_docs` |

## Recommended Workflow

1. **Detect source type** from the user's input
2. **Generate or fetch config** using `generate_config` or `fetch_config` if needed
3. **Estimate scope** with `estimate_pages` for documentation sites
4. **Scrape the source** using the appropriate scraping tool
5. **Enhance** with `enhance_skill` if the user wants AI-powered improvements
6. **Package** with `package_skill` for the target platform
7. **Export to vector DB** if requested using `export_to_*` tools

## Available MCP Tools

### Config Management
- `generate_config` — Generate a scraping config from a URL
- `list_configs` — List available preset configs
- `validate_config` — Validate a config file

### Scraping (use based on source type)
- `scrape_docs` — Documentation sites
- `scrape_github` — GitHub repositories
- `scrape_pdf` — PDF files
- `scrape_video` — Video transcripts
- `scrape_codebase` — Local code analysis
- `scrape_generic` — Jupyter, HTML, OpenAPI, AsciiDoc, PPTX, RSS, manpage, Confluence, Notion, chat

### Post-processing
- `enhance_skill` — AI-powered skill enhancement
- `package_skill` — Package for target platform
- `upload_skill` — Upload to platform API
- `install_skill` — End-to-end install workflow

### Advanced
- `detect_patterns` — Design pattern detection in code
- `extract_test_examples` — Extract usage examples from tests
- `build_how_to_guides` — Generate how-to guides from tests
- `split_config` — Split large configs into focused skills
- `export_to_weaviate`, `export_to_chroma`, `export_to_faiss`, `export_to_qdrant` — Vector DB export

## Commands

```bash
# Detect source type from input
#   URL → scrape_docs
#   owner/repo → scrape_github
#   file.pdf → scrape_pdf
#   video URL → scrape_video

# Basic workflow pipeline
python3 -c "
from skill_seekers import scrape_docs, enhance_skill, package_skill
# Step 1: Scrape
config = scrape_docs('https://example.com/docs', output_dir='./scraped')
# Step 2: Enhance
enhanced = enhance_skill(config, target_platform='hermes')
# Step 3: Package
package_skill(enhanced, format='SKILL.md', output='./output/')
"
```

## Dependencies

- Python 3.10+
- `skill-seekers` package (tools for scraping and packaging)
- Internet access for documentation scraping
- API keys for vector DB exports (optional)

## Process

1. **Design** — Define interface, identify patterns, plan implementation
1. **Implement** — Write code following existing conventions, add tests
1. **Verify** — Run tests, check integration, validate behavior

## Verification

- [ ] Skill has SKILL.md with frontmatter
- [ ] Has ## When to Use with triggers
- [ ] Has ## Process section
- [ ] Has ## Verification checklist
- [ ] Lint passes with 0 errors