Skip to content
AI workflow

We went from plain text to SPAs and back to plain text — and the agents are the reason

The internet came full circle. It started as plain HTML with hyperlinks. Then it bloated into single-page React applications. Now it is collapsing back to plain text markdown with hyperlinks. The difference is who is reading. First time around, humans. This time, AI agents.

Based on observations about the shift from visual web interfaces to agent-consumable markdown formats in 2025-2026.

The arc nobody planned

The web started as plain text. Tim Berners-Lee wrote HTML documents with hyperlinks and that was it. No build step, no bundler, no hydration. Just text pointing to other text.

Then we needed interactivity. jQuery gave us dropdowns. Angular gave us SPAs. React gave us a virtual DOM that diffed against itself so we could render a todo list. We added thousands of lines of JavaScript so humans could click buttons that looked nice. The browser became an operating system running inside an operating system.

Now the primary consumer of web content is shifting from humans to agents. And agents do not care about your animations, your hover states, or your client-side routing. They care about structured text. So the output format is collapsing back to markdown.

Plain text to bloated SPAs to plain text. Full circle.

The agent infrastructure layer

MCPs — Model Context Protocols — are the clearest signal that the web is being rebuilt for a different reader. An MCP is not a human interface. It is a natural language API for bots to run commands on our behalf. You expose tools, the agent calls them, things happen. No pixels involved.

Stainless.com takes this a step further. They auto-generate SDKs and MCPs from OpenAPI specs. You have an OpenAPI JSON file describing your API, and Stainless turns that into a fully typed SDK plus an MCP server — for free. The entire pipeline from API spec to agent-consumable interface is automated. No frontend developer needed.

This is infrastructure for a world where agents are the primary users of your product. If your service has an API, it can have an MCP. If it has an MCP, agents can use it. The interface layer is plain text all the way down.

Why I built allmd

I built allmd because I kept hitting the same wall. I needed to feed content into an agent, but the content was trapped in formats agents cannot read. A PDF. A YouTube video. A webpage drowning in JavaScript.

allmd converts any URL, PDF, video, or audio file into clean markdown. Because agents need markdown, not pixels. That is the entire product thesis in one sentence. The web is full of information locked behind visual rendering, and agents need that information as structured text.

The irony is thick. React developers — people who spent careers building rich visual interfaces — are now building tools that output markdown for agents to consume. The framework that dominated the last decade of frontend development is producing tools for a world that does not need a frontend.

The web is being re-read

Every page, every doc, every API is being flattened back to text. The entire web is being re-read by machines. Not crawled for search ranking like Google did. Read for comprehension. Agents are consuming documentation, parsing changelogs, reading blog posts, and executing tasks based on what they find.

This changes what matters about your content. Visual hierarchy was for humans scanning a page. Structured text is for agents parsing information. The heading levels, the bold text, the link labels — those are not style choices anymore. They are semantic signals for a non-human reader.

What this means for builders

If your product cannot be represented as structured text, agents cannot use it. That is not a theoretical concern. It is a practical constraint that is already shaping what gets adopted.

This is not a regression. It is a different reader with different needs. Humans needed visual hierarchy, whitespace, colour, animation — all the things that make information scannable by eyes. Agents need structured text, consistent formatting, clear labels — all the things that make information parseable by models.

The builders who understand this are already shipping. The ones still debating which React framework to use for their dashboard are building for a reader that is becoming less important by the month.

The web started as text for humans. It is becoming text for agents. Same format, different audience, entirely different implications.

Keep reading