Web Summary Skill

URL content summarization and topic extraction via FastMCP, using Z.AI service for LLM processing.

When to Use This Skill

• Summarize web articles, blog posts, documentation
• Extract and summarize YouTube video transcripts
• Generate semantic topic links for note-taking (Logseq, Obsidian)
• Create short, medium, or long summaries
• Extract key concepts from text

Architecture

┌─────────────────────────────┐
│  Claude Code / VisionFlow   │
│  (MCP Client)               │
└──────────────┬──────────────┘
               │ MCP Protocol (stdio)
               ▼
┌─────────────────────────────┐
│  Web Summary MCP Server     │
│  (FastMCP - Python only)    │
└──────────────┬──────────────┘
               │ HTTP (port 9600)
               ▼
┌─────────────────────────────┐
│  Z.AI Service               │
│  (Cost-effective Claude)    │
└─────────────────────────────┘

Tools

Examples

# Summarize a web article
summarize_url({
    "url": "https://example.com/article",
    "length": "medium",
    "include_topics": True,
    "format": "logseq"
})

# Get YouTube transcript
youtube_transcript({
    "video_id": "dQw4w9WgXcQ",  # or full URL
    "language": "en"
})

# Generate topic links
generate_topics({
    "text": "Your text content here...",
    "max_topics": 10,
    "format": "obsidian"
})

Output Formats

Logseq

- [[Topic One]]
- [[Topic Two]]
- [[Machine Learning]]

Obsidian

- [[Topic One]]
- [[Topic Two]]
- [[Machine Learning]]

Plain

- Topic One
- Topic Two
- Machine Learning

Environment Variables

Troubleshooting

Z.AI connection failed:

# Check Z.AI service status
supervisorctl status claude-zai

# Test Z.AI directly
curl -X POST http://localhost:9600/chat \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Hello"}'

VisionFlow Integration

This skill exposes web-summary://capabilities resource for discovery by VisionFlow's MCP TCP client on port 9500.