Make: Makezine Maker Project Scraper

Extract structured metadata from all curated maker projects published on Makezine.com — the flagship publication for DIY makers, hackers, and craft enthusiasts.

What it does

Crawls all ~2,600 project pages across Makezine's dedicated project sitemaps and extracts per-project metadata: title, author, category, tags, tools list, materials list, excerpt, publish date, thumbnail, and project URL. No subscription required — all metadata is publicly accessible.

Output fields

Field	Type	Description
project_id	string	Stable slug-based project identifier
title	string	Project title
author	string	Author name
slug	string	URL slug
category	string	Project category (from articleSection, e.g. "Woodworking, Workshop")
tags	string[]	Keywords / tags
difficulty	string	Difficulty level (null — not structured on this site)
time_required	string	Time estimate (null — not structured on this site)
tools_summary	string[]	List of tools required
materials_summary	string[]	List of materials/parts required
excerpt	string	Short project description
published_at	string	ISO 8601 publish date
thumbnail_url	string	Project thumbnail image URL
project_url	string	Canonical project page URL

Use cases

• Maker content datasets — Train or augment LLMs / recommendation systems on curated DIY project content
• Research — Analyze maker trends, popular tools, category distributions across 15+ years of publications
• Instructables complement — Pair with the Instructables scraper for a curated + UGC maker dataset
• SEO / content analysis — Track which categories and tags dominate Make: magazine's editorial catalog

Input

Parameter	Type	Default	Description
maxItems	integer	10	Maximum number of project records to scrape. Set to 0 for all ~2,600 projects.

Notes

• Covers woodworking, electronics, craft, science, photography, food, furniture, and more
• Tools and materials lists may be empty on older projects that predate the structured sidebar format
• difficulty and time_required are not present as machine-readable fields on Makezine — both return null
• Full corpus run (~2,600 projects) takes approximately 10 minutes at default concurrency