- Home
- Skills
- Productivity
- web2md — Turn Any Web Article into Clean Local Markdown
web2md — Turn Any Web Article into Clean Local Markdown
Convert web articles into clean, self-contained local Markdown files — article body only, images downloaded locally, YAML frontmatter included. One command, works as an AI agent skill or a plain CLI tool.
Free
Works with the AI Agents you already use
web2md — Turn Any Web Article into Clean Local Markdown
Example session with this skill installed
Convert this web article into a clean, self-contained local Markdown file — article body only, images downloaded locally, with YAML frontmatter included. Save it so I can read it offline in Obsidian.
- Read your context and instructions
- Compiled the web2md
Saved: article.md + images/
title: Example Web Article
source: https://example.com/article
Example Web Article
The article body is preserved here, with navigation, ads, sidebars, related-post lists and comment sections removed. Code blocks and inline formatting are kept intact.
The downloaded image is referenced with a relative path, so the file reads fully offline.
Files created
- article.md
- images/figure-1.png
Next step: drop article.md into Obsidian or any notes vault — no further cleanup is needed.
Connects securely to your tools. The creator never sees your data.
About this skill
web2md fetches a web article and turns it into a tidy local .md file plus an images/ folder. Navigation, ads, sidebars, related-post lists and comment sections are stripped automatically; images (including lazy-loaded ones hidden in data-src) are downloaded and referenced with relative paths, so the result is fully offline-readable and can be dropped straight into Obsidian or any notes vault. How it finds the article: every candidate container is scored by paragraph text volume discounted by link density, then boilerplate blocks (breadcrumb / comments / related / share / nav …) are removed with a safety threshold so real content is never cut. If auto-detection ever misses, a --selector flag lets you name the content container directly. Verified on WordPress, Blogger and English blogs (Chinese and English content, code blocks and inline formatting preserved). Compared with readability/trafilatura libraries, web2md is a single-file, zero-config command that also handles image localization out of the box — built to be handed to a coding agent as a skill. Requirements: Python 3.8+ and lxml (pip install lxml). MIT licensed. Also free on GitHub: https://github.com/oldxianyu/web2md
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 13 days ago
- Free with an Agensi account
Requires