6 min read

Monitor Website Changes with Python

Get notified when any website changes. Clean Markdown diff + Claude for intelligent detection. No CSS selectors. Complete Python implementation with SQLite, diff generation, and Slack/email alerts.

Read article
7 min read

Turn Any Website Into a JSON API with Python

Most websites don't have an API. Here's how to build one yourself in 30 lines: fetch clean Markdown with UnWeb, extract structured data with Claude, wrap it in FastAPI.

Read article
8 min read

Building a Competitor Price Monitor in Python

Track competitor pricing automatically: fetch product pages as clean Markdown, extract prices with Claude, store history, and get notified when something changes. No selectors to maintain.

Read article
9 min read

Extracting Structured Data from Any Web Page with Python and Claude

CSS selectors break on site changes. LLMs need clean input. Here's the approach that works: convert any URL to clean Markdown with UnWeb, then use Claude to extract typed JSON from the text.

Read article
8 min read

How to Scrape JavaScript-Rendered Pages in Python

requests.get() returns skeleton HTML on JavaScript-rendered sites. Here's why that happens and three ways to fix it — including why headless browsers aren't always the answer.

Read article
4 min read

Web Scraping with Node.js: Clean Markdown from Any URL

axios + cheerio fails silently on React, Next.js, and Vue sites. Here's a quality-gated approach using the UnWeb Node.js SDK.

Read article
7 min read

Building a Web Research Agent with Python and Claude

A web research agent that actually works: fetch any URL, convert to clean Markdown, extract structured facts with Claude, and synthesize a report. No skeleton HTML, no prompt engineering for messy DOM.

Read article
8 min read

Markdown as Developer Workflow Infrastructure

Markdown is now the universal exchange format for LLM pipelines, AI agents, and developer tooling. Here's how to treat web-to-Markdown conversion as first-class infrastructure rather than an afterthought.

Read article
8 min read

How to Build a RAG Pipeline with Live Web Data Using Python

Stop polluting your vector store with skeleton HTML. Use UnWeb's quality score to pre-filter JS-rendered pages before they corrupt your RAG pipeline.

Read article
4 min read

How to Convert Any Webpage to Markdown from Claude Code with UnWeb MCP

Install the UnWeb MCP server in 3 lines of config. Your AI coding assistant can convert any URL to clean Markdown or crawl entire doc sites — with a quality score on every response.

Read article