--- title: Markdown url: https://ftrlg.com/en/wiki/markdown/ lang: en published: 2026-10-13 version: 1 summary: A simple plain-text markup designed to go unnoticed by people, which became a common language for humans and AI. cite_as: "future.log, «Markdown», https://ftrlg.com/en/wiki/markdown/" license: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ --- # Markdown **Markdown** is a way of marking up plain text with simple characters: `#` for a heading, `*` for a list item, `**` for bold. It was designed so that people would not notice the markup. Twenty years later it turned out to suit machines just as well: chatbots answer in Markdown, instructions for [agents](https://ftrlg.com/en/wiki/autonomous-ai-agents/) are written in it, and websites serve pages to AI in it. ## Markup you can ignore The idea is older than Markdown. On the first wiki, Ward Cunningham's WikiWikiWeb (1995), text was marked up with ordinary characters: two apostrophes for italics, three for bold, an asterisk for a list item.[^1][^2] Wikipedia is still written the same way.[^3] In 2004 John Gruber turned this idea into a format for bloggers: Markdown, an “easy-to-read, easy-to-write plain text format”.[^4] He modeled it on the way people already format their emails. The main rule: marked-up text should read like ordinary text, “without looking like it's been marked up with tags”.[^5] The format spread faster than anyone could pin it down, and every service grew its own flavor. In 2014 Jeff Atwood and co-authors merged them into a common specification, CommonMark;[^6][^7] in 2017 GitHub, where it is used to write project descriptions, built its own Markdown on top of it.[^8] That is how the bloggers' format became the language of programmers. ## When the reader is a model A language model needs the same thing a person does: to see the structure of a text without extra noise. An HTML page is almost entirely service markup, and a model pays for every token. By Cloudflare's count, the same post costs 16,180 tokens in HTML and 3,150 in Markdown, 80% less.[^9] This is part of why Markdown became the format in which AI and people exchange text in both directions. **AI to human.** [OpenAI](https://ftrlg.com/en/models/openai/) states it plainly in its rules for model behavior: unless otherwise specified, responses are formatted in Markdown.[^10] Hence the headings, lists and tables in chatbot answers. **Human to agent.** For coding agents, project rules go into a Markdown file at the root of the repository. The standard for this became AGENTS.md, “a README for agents”, which OpenAI released in 2025; by December more than 60,000 open-source projects used it, and it is now governed by the Agentic AI Foundation.[^11][^12] **Website to agent.** In 2024 Jeremy Howard proposed putting an llms.txt file on websites: a table of contents for models in Markdown, because it is “the most widely and easily understood format for language models”.[^13] In 2026 Cloudflare began serving agents Markdown instead of HTML on request, calling it “the lingua franca for agents”.[^9] ## Wikipedia: same idea, different path Wikipedia is older than Markdown and is written in its own wiki markup, of the same kind as WikiWikiWeb's.[^3] But it is not served to machines in Markdown: in 2025 Wikimedia Enterprise released articles for machine learning in JSON, “instead of scraping or parsing raw article text”.[^14] Converting Wikipedia to Markdown for models is so far left to enthusiasts: Santhosh Thottingal, an engineer at the Wikimedia Foundation, wrote such a program as a personal project, because otherwise the service markup has to be stripped out of HTML for models.[^15] So the largest wiki, heir to the idea of simple markup, speaks a different language with AI: the language of structured data (see [Wikipedia after AI](https://ftrlg.com/en/wiki/wikipedia-after-ai/)). ## Where Markdown gets in the way Markup is not neutral. In a 2024 study, the quality of GPT-3.5-turbo's code translation varied by up to 40% depending on whether the prompt was given as plain text, Markdown, JSON or YAML.[^16] And when a model's answer is read by a program rather than a person, asterisks and hash signs only get in the way: since late 2024, OpenAI's reasoning models in the API answer without Markdown by default, and developers turn it on with a special string, “Formatting re-enabled”.[^17] > **[EDITORS' CONCLUSION]** Markdown won not because it is better than other markup languages, but because it works equally well for two readers: it does not get in a person's way, and for a model it costs several times less than HTML. A format invented for bloggers became the interface between people and AI, and where the only reader left is a machine, it is already being displaced by strict formats such as JSON. ## See also [Wikipedia after AI](https://ftrlg.com/en/wiki/wikipedia-after-ai/) · [Autonomous AI agents](https://ftrlg.com/en/wiki/autonomous-ai-agents/) · [Large language model](https://ftrlg.com/en/wiki/large-language-model/) · [Open source](https://ftrlg.com/en/wiki/open-source/) ## References [^1]: WikiHistory — WikiWikiWeb (c2.com). [wiki.c2.com](https://wiki.c2.com/?WikiHistory) [^2]: TextFormattingRules — WikiWikiWeb (c2.com). [wiki.c2.com](https://wiki.c2.com/?TextFormattingRules) [^3]: Help:Formatting — MediaWiki. [mediawiki.org](https://www.mediawiki.org/wiki/Help:Formatting) [^4]: Markdown — Daring Fireball. [daringfireball.net](https://daringfireball.net/projects/markdown/) [^5]: Markdown: Syntax — Daring Fireball. [daringfireball.net](https://daringfireball.net/projects/markdown/syntax) [^6]: Standard Flavored Markdown — Coding Horror, 03.09.2014. [blog.codinghorror.com](https://blog.codinghorror.com/standard-flavored-markdown/) [^7]: Standard Markdown is now Common Markdown — Coding Horror, 04.09.2014. [blog.codinghorror.com](https://blog.codinghorror.com/standard-markdown-is-now-common-markdown/) [^8]: A formal spec for GitHub Flavored Markdown — GitHub Blog, 14.03.2017. [github.blog](https://github.blog/engineering/user-experience/a-formal-spec-for-github-markdown/) [^9]: Introducing Markdown for Agents — Cloudflare Blog, 12.02.2026. [blog.cloudflare.com](https://blog.cloudflare.com/markdown-for-agents/) [^10]: OpenAI Model Spec — OpenAI. [model-spec.openai.com](https://model-spec.openai.com/2026-08-18.html) [^11]: Linux Foundation Announces the Formation of the Agentic AI Foundation (AAIF) — Linux Foundation, 09.12.2025. [linuxfoundation.org](https://www.linuxfoundation.org/press/linux-foundation-announces-the-formation-of-the-agentic-ai-foundation) [^12]: AGENTS.md. [agents.md](https://agents.md/) [^13]: The /llms.txt file — Jeremy Howard, 03.09.2024. [llmstxt.org](https://llmstxt.org/) [^14]: Wikipedia Kaggle Dataset using Structured Contents Snapshot — Wikimedia Enterprise, 16.04.2025. [enterprise.wikimedia.com](https://enterprise.wikimedia.com/blog/kaggle-dataset/) [^15]: CLI for transforming Wikipedia articles to text, markdown, and JSON — Santhosh Thottingal, 14.03.2026. [thottingal.in](https://thottingal.in/blog/2026/03/14/wikipedia-article-transform/) [^16]: J. He et al. Does Prompt Formatting Have Any Impact on LLM Performance? — arXiv, 15.11.2024. [arxiv.org](https://arxiv.org/abs/2411.10541) [^17]: Reasoning best practices — OpenAI API docs. [developers.openai.com](https://developers.openai.com/api/docs/guides/reasoning-best-practices) — future.log, «Markdown», https://ftrlg.com/en/wiki/markdown/ · CC BY 4.0