Defuddle

The preferred method for reading web content. Returns clean Markdown via the defuddle.md service.

When to use

Use defuddle first for any URL content fetching. It produces cleaner output than WebFetch for most sites because it extracts the main content and strips navigation, ads, and boilerplate.

• Any URL the user asks you to read — articles, blog posts, docs, tweets, forum threads
• Any URL you need to fetch yourself — documentation lookups, reference pages, linked resources
• Only fall back to WebFetch if defuddle returns empty or errors out

How it works

defuddle.md is a public service that extracts the main content from a webpage and returns it as Markdown with YAML frontmatter.

curl -sL "https://defuddle.md/<url>"

The <url> is the target URL without the protocol (no https://).

Examples

Fetch an article:

curl -sL "https://defuddle.md/example.com/blog/some-post"

Fetch a tweet / X post:

curl -sL "https://defuddle.md/x.com/username/status/123456789"

Usage notes

• Always use -sL flags (silent mode + follow redirects)
• The request URL must use https:// prefix: https://defuddle.md/...
• Strip https:// or http:// from the target URL before appending
• The response is Markdown text — you can use it directly or extract specific sections
• If the response is empty or an error page, the target site may not be supported; fall back to WebFetch or inform the user