tapestry

Tapestry: Unified Content → Action Workflow

This skill combines content extraction with actionable planning - turning any learning resource (YouTube videos, articles, PDFs) into Ship-Learn-Next action plans in one seamless flow.

When to Use This Skill

Activate when the user:

• Says "tapestry [URL]" or "weave [URL]"
• Wants to "extract and plan from [URL]"
• Asks to "make [URL] actionable"
• Says "help me implement [URL]"
• Provides a URL and wants both content + action plan
• Wants to turn learning content into concrete steps

Core Philosophy

Tapestry weaves three threads together:

1. Extract - Get clean content from any source
2. Synthesize - Identify actionable lessons
3. Plan - Create Ship-Learn-Next cycles

Result: From URL to shippable action plan in one flow.

How Tapestry Works

Step 1: Detect Content Type

When user provides a URL, automatically detect:

# YouTube detection
if [[ "$URL" =~ youtube\.com|youtu\.be ]]; then
    CONTENT_TYPE="youtube"
# Article/blog detection
elif [[ "$URL" =~ ^https?:// ]]; then
    CONTENT_TYPE="article"
# PDF detection (if URL ends in .pdf)
elif [[ "$URL" =~ \.pdf$ ]]; then
    CONTENT_TYPE="pdf"
fi

Step 2: Extract Content (Automatic)

Based on content type, use the appropriate extraction method:

For YouTube Videos

Use the youtube-transcript skill:

# Activate youtube-transcript skill
# This handles:
# - Installation check (yt-dlp)
# - Subtitle detection (manual → auto-generated → Whisper)
# - VTT to plain text conversion
# - Deduplication

Result: Clean transcript saved as [Video Title].txt

For Articles/Blogs

Use the article-extractor skill:

# Activate article-extractor skill
# This handles:
# - Tool detection (reader/trafilatura/fallback)
# - Content extraction
# - Clutter removal
# - Clean text output

Result: Clean article saved as [Article Title].txt

For PDFs

Direct extraction:

# Check for PDF tools
if command -v pdftotext &> /dev/null; then
    pdftotext "$PDF_URL" output.txt
elif command -v mutool &> /dev/null; then
    mutool draw -F txt -o output.txt "$PDF_URL"
else
    echo "PDF extraction requires pdftotext or mutool"
    echo "Install: brew install poppler (macOS) or apt install poppler-utils (Linux)"
fi