Blog
Scraping guides, MCP tutorials, and opinionated takes from the Stekpad team. We write about what we ship.
Why your AI agent needs MCP-native scraping (2026)
Most "AI scrapers" were built for humans clicking buttons. The world changed. Here's what a scraper designed for AI agents looks like.
Your AI agent needs live data, not yesterday's CSVs
Why scheduled cron scrapers can't feed AI agents, and how MCP changes the contract between the model and the web.
The authenticated scraping guide for 2026
How to scrape pages that require a login without leaking cookies to a server. A practical guide using the Stekpad cookie bridge.
Beyond cron jobs — why batch scraping can't feed a live product
Cron-scheduled scrapers made sense when the consumer was a CSV. They don't when the consumer is a Claude session.
Build a lead enrichment pipeline with Claude and Stekpad
A step-by-step tutorial — from raw URLs to enriched rows — using Claude Desktop, the Stekpad MCP server, and 19 native enrichers.
A Clay alternative for teams that want to own their data
How Stekpad compares to Clay for enrichment workflows — and when each tool is the better fit.
How to scrape Google Maps leads without getting blocked
A practical walkthrough of scraping Google Maps for local business leads using the Stekpad extract verb.
How to scrape LinkedIn safely in 2026
The honest guide to scraping LinkedIn profiles and company pages without burning your session or violating ToS boundaries that matter.
MCP explained for growth teams (no PhD required)
What MCP actually is, why it matters for growth ops, and how to install the Stekpad MCP server in Claude Desktop in under a minute.
A PhantomBuster alternative for 2026
How Stekpad compares to PhantomBuster for automation workflows — honest trade-offs, pricing, and when each tool wins.
Why we replaced Zapier with an MCP server for scraping
Zapier broke the moment our scraper needed to think. Here is what we rebuilt — and why MCP was the right shape for the job.
How to scrape any page straight into Google Sheets
A tutorial for piping Stekpad scrapes into Google Sheets using the native export, with a live example and a credit cost breakdown.
Scrape without XPath — what a schema-first extractor actually looks like
XPath selectors break every time a site redesigns. Schema-first extraction with an AI selector doesn't. Here is why.
Is web scraping legal in 2026? A practical guide
What the 2026 case law actually says about scraping public data, ToS boundaries, and the cookie bridge approach.
Why every existing scraper is broken for AI agents
Legacy scrapers were built for analysts with spreadsheets. Agents need a different contract — stateless, typed, priced per call.
Why Stekpad runs authenticated scrapes in your browser
The cookie bridge architecture — why we chose to fetch pages in the user's Chrome instead of proxying cookies to a server.
Building AI agents that scrape — an MCP-first guide
An end-to-end guide to wiring an AI agent to the Stekpad MCP server, with working tool configs for Claude Desktop and Cursor.
19 native enrichers vs third-party vendors — a case for owning the stack
Why Stekpad ships 19 enrichers in the box instead of reselling third-party APIs — and what that means for your data, pricing, and trust.
The cookie bridge — architecture deep dive
How Stekpad scrapes pages that require a login without ever storing a single cookie on our servers. Diagrams, sequence flows, and the exact trust boundary.
Firecrawl vs Apify vs Stekpad — a 2026 technical comparison
An honest, feature-by-feature comparison of the three scraping stacks builders pick between in 2026. With a "when they win" section for each.
The 5 API verbs that replaced my scraping stack
Scrape, crawl, map, extract, search — five verbs that replaced a dozen tools in our own pipeline. What each one does and when to call it.