defuddle

Defuddle

Use Defuddle CLI to extract clean readable content from web pages. Prefer over WebFetch for standard web pages — it removes navigation, ads, and clutter, reducing token usage.

Preferred Tool Order

1. Use Defuddle for standard article-style pages where clean markdown extraction is the main goal.
2. Use fetch when Defuddle is unavailable, the page is not article-like, or raw retrieval is preferable.
3. Use imagefetch when the task depends on understanding page images as well as text.

Availability Check

Before using Defuddle, verify the CLI is installed:

defuddle --help

• Exit code 0: Defuddle is available.
• Non-zero or command-not-found: Defuddle is unavailable. Use the fallback strategy below.

If Defuddle is unavailable, the page is not a standard article page, or you need raw page retrieval rather than article extraction, fall back to the fetch MCP when it is available. If the task depends on understanding page images, prefer imagefetch over plain fetch.

If not installed: npm install -g defuddle

Fallback Strategy

• Use fetch for raw or general page retrieval when Defuddle is unavailable.
• Use imagefetch instead of plain fetch when page imagery is part of the task.
• If the page is highly interactive rather than article-like, switch to a browser-oriented tool instead of forcing Defuddle.

No-Tool Behavior

If neither Defuddle nor the documented MCP fallback is available, say that clean page extraction is unavailable in the current runtime.

• Ask the user for the source URL or page content if a manual path is still viable.
• Do not describe the extraction as complete when the tool path was missing.

Usage

Always use --md for markdown output:

defuddle parse <url> --md

Save to file:

defuddle parse <url> --md -o content.md

Extract specific metadata:

defuddle parse <url> -p title
defuddle parse <url> -p description
defuddle parse <url> -p domain

Output formats