Weaver Demo

About

Sun, 01 Feb 2026 00:00:00 +0000

This is a demo site for the Weaver Hugo theme, built by Brandon High.

The content and layout of this site were generated by Claude Code using Claude Sonnet 4.5, Anthropic’s agentic coding tool. From the blog posts to the theme templates, the site was built through conversation: a human describing what they wanted, and an AI writing the code and copy to make it happen.

It’s a small example of what the indie bot web can look like.

Build Your Own

Sun, 01 Feb 2026 00:00:00 +0000

Paste this into your agentic coding tool of choice to set up a personal blog deployed to GitHub Pages.

What you’ll get

A Hugo static site using the Weaver theme (IndieWeb-friendly)
GitHub Pages deployment via GitHub Actions
A ready-to-write blog with posts, an about page, and navigation

Prerequisites

Git installed
A GitHub account with a repo created (can be empty)
Hugo extended installed locally (for previewing)
GitHub CLI (gh) installed and authenticated

Prompt

Copy everything below this line and paste it into your LLM coding tool, replacing the placeholder values with your own.

Learning Guide

Sun, 01 Feb 2026 00:00:00 +0000

If you used the Build Your Own prompt to set up your site, a lot just happened very quickly. You now have a working website, but you might not understand how or why it works. That’s completely okay. You don’t need to understand everything at once, or ever, honestly. Plenty of people drive cars without knowing how engines work.

But if you’re curious, this guide walks through the technologies your site uses, starting from the basics. Take it at your own pace.

Limitations of LLMs

Sun, 01 Feb 2026 00:00:00 +0000

If you used an AI tool to build your site, it’s worth understanding what that tool is bad at. Not to scare you off (you’ve already got a working website, which is great) but because knowing the limits helps you use it better.

They reflect their training data, including its biases

LLMs are trained on enormous amounts of text from the internet. That text contains biases: cultural, commercial, linguistic, and structural. The AI doesn’t “choose” to be biased; it just pattern-matches against what it’s seen, and what it’s seen is not neutral.

The Birth of the Web

Sun, 01 Feb 2026 00:00:00 +0000

In 1989, a British computer scientist named Tim Berners-Lee submitted a proposal to his managers at CERN titled “Information Management: A Proposal.” His supervisor famously scribbled “Vague but exciting” on the cover. That understated reaction belied the enormity of what was to come.

The Problem

CERN was drowning in information. Thousands of researchers from around the world collaborated on experiments, generating massive amounts of documentation spread across incompatible systems. Finding and linking related information was a constant struggle. Berners-Lee saw this as a universal problem, not just at CERN, but everywhere people needed to share knowledge.

The Indie Bot Web

Sun, 01 Feb 2026 00:00:00 +0000

In my last post, I wrote about how Tim Berners-Lee created the World Wide Web and, crucially, gave it away for free. That openness unleashed a wave of creativity, but over the decades, actually building for the web became increasingly complex. What once required just a text editor and an FTP client eventually demanded familiarity with frameworks, build tools, hosting platforms, DNS configuration, and more. The web was free, but the expertise to use it was not evenly distributed.