url-to-markdown（URL TO Markdown 控制）

URL to Markdown

Fetches any URL via Chrome CDP and converts HTML to clean markdown.

Script Directory

Important: All scripts are located in scripts/ subdirectory of this skill.

Agent Execution Instructions:

1. Determine this SKILL.md file's directory path as SKILL_DIR
2. Script path = ${SKILL_DIR}/scripts/<script-name>.ts
3. Replace all ${SKILL_DIR} in this document with actual path

Script Reference:

Features

• Chrome CDP for full JavaScript rendering
• Two capture modes: auto or wait-for-user
• Clean markdown output with metadata
• Handles login-required pages via wait mode

Usage

# Auto mode (default) - capture when page loads
npx -y bun ${SKILL_DIR}/scripts/main.ts <url>

# Wait mode - wait for user signal before capture
npx -y bun ${SKILL_DIR}/scripts/main.ts <url> --wait

# Save to specific file
npx -y bun ${SKILL_DIR}/scripts/main.ts <url> -o output.md

Options

Capture Modes

Auto Mode (default)

Page loads → waits for network idle → captures immediately.

Best for:

• Public pages
• Static content
• No login required

Wait Mode (--wait)

Page opens → user can interact (login, scroll, etc.) → user signals ready → captures.

Best for:

• Login-required pages
• Dynamic content needing interaction
• Pages with lazy loading

Agent workflow for wait mode:

1. Run script with --wait flag
2. Script outputs: Page opened. Press Enter when ready to capture...
3. Use AskUserQuestion to ask user if page is ready
4. When user confirms, send newline to stdin to trigger capture

Output Format

---
url: https://example.com/page
title: "Page Title"
description: "Meta description if available"
author: "Author if available"
published: "2024-01-01"
captured_at: "2024-01-15T10:30:00Z"
---

# Page Title

Converted markdown content...

Mode Selection Guide

When user requests URL capture, help select appropriate mode:

Suggest Auto Mode when:

• URL is public (no login wall visible)
• Content appears static
• User doesn't mention login requirements

Suggest Wait Mode when:

• User mentions needing to log in
• Site known to require authentication
• User wants to scroll/interact before capture
• Content is behind paywall

Ask user when unclear:

The page may require login or interaction before capturing.

Which mode should I use?
1. Auto - Capture immediately when loaded
2. Wait - Wait for you to interact first

Output Directory

Each capture creates a file organized by domain:

url-to-markdown/
└── <domain>/
    └── <slug>.md

Path Components:

• <domain>: Site domain (e.g., example.com, github.com)
• <slug>: Generated from page title or URL path (kebab-case)

Slug Generation:

1. Extract from page title (preferred) or URL path
2. Convert to kebab-case, 2-6 words
3. Example: "Getting Started with React" → getting-started-with-react

Conflict Resolution: If url-to-markdown/<domain>/<slug>.md already exists:

• Append timestamp: <slug>-YYYYMMDD-HHMMSS.md
• Example: getting-started.md exists → getting-started-20260118-143052.md

Error Handling

Environment Variables

Extension Support

Custom configurations via EXTEND.md.

Check paths (priority order):

1. .content-gen-skills/url-to-markdown/EXTEND.md (project)
2. ~/.content-gen-skills/url-to-markdown/EXTEND.md (user)

If found, load before workflow. Extension content overrides defaults.