mozilla/fathom

Health v2 · maintenance only

Flags: no_releases archived

Adoption not part of the score

1963 stars · 73 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Fathom is a supervised-learning framework for recognizing and classifying parts of web pages, tagging DOM nodes with types and probabilities using a Prolog-like rule language and a neural-net-based trainer. It is now deprecated and unmaintained.

Use cases

• detect pop-ups on web pages
• extract address forms from html
• classify web pages by type
• label a corpus of web pages for training
• recognize login and password forms
• extract structured meaning from dom nodes

When to choose

• you need rule-based plus trained semantic extraction from DOM trees
• you want to label and train on a corpus of labeled web pages
• you need probabilistic tagging of page elements like forms or pop-ups

When to avoid

• you need an actively maintained project
• you want general web scraping without a training workflow
• you need production-grade support or timely bug fixes

Facets

framework · maturity abandoned

machine-learning web-scraping parser browser-extension machine-learning web-development browser browser-extension supervised-learning dom-extraction prolog-like-rules deprecated web-page-classification fathomfox natural-language-processing javascript nodejs

Member repositories

For agents

markdown · JSON · MCP: product_card(name="mozilla/fathom")