How to Write a Good llms.txt File (2026)
An llms.txt file is a routing page for coding agents, not a copy of your documentation. Where to put it, how to group the links, and what to check before you ship it.
30 stories
An llms.txt file is a routing page for coding agents, not a copy of your documentation. Where to put it, how to group the links, and what to check before you ship it.
Documentation took 55% of page fetches across 500 tracked coding-agent runs. Seven requirements for docs an agent can finish a real task with, plus a method for testing them the way an agent would use them.
Absolutely yes. Coding agents use llms.txt to understand your product and navigate your documentation, and our research shows how often they read it.
Claude Code uses two separate web pipelines: a server-side search that returns links and findings, and a local fetch that sends a page through Markdown conversion and usually a Haiku extraction pass.
Open models show similar preferences compared to Claude and Codex across 1,000+ sessions. That makes them useful for efficiently finding opportunities.
Agent Experience is the practice of making products easy for AI agents to discover, understand, use, and recover with. Learn how to measure AX and improve docs, onboarding, errors, URLs, and APIs.
Agent Preference Optimization is how companies measure and improve whether coding agents choose their product. Here is how to establish a baseline, understand why competitors win, act on the evidence, and track the result.
Coding agents opened llms.txt in 36.3% of build sessions, but only 0.5% of vendor-selection sessions. The task determines whether the file matters.
Server logs showed no ChatGPT-User requests for llms.txt. Instrumented sessions showed exactly where those reads went.
Humans should decide what content says, AI should write most of the prose, and humans should review the result.
Serve a Markdown version of every docs URL so agents get cleaner context, current package versions, exact install commands, and a map of your docs.
Coding agents now choose and install the tools developers build with. Agent Led Growth is the channel that decides which tool wins, and it has two halves: get recommended, then get implemented properly.
How Natalia Amorim used Gauge to Increase Posthog LLM Traffic 41x.
Grok mentions more than double than brands Perplexity does while OpenAI and Anthropic land in the middle.
How Braintrust's Head of Growth Marketing, Raman Hundal, built an AI-search content engine from scratch and grew their visibility from 2.5% to 45% of AI answers in their space.
ChatGPT ads are live and self-serve. Here's what to track, how they differ from Google Ads, and how Gauge connects paid performance to organic AI visibility via MCP.
Understanding the different forms of content, and why first party content is the most effective strategy for GEO.
Complete guide to understanding how AI speaks about your brand and how to optimize for it.
Showing 18 of 30 stories
Load more storiesGet the top AI news and tools straight to your inbox every week.
Weekly · AI news
Join 15,000+ AI enthusiasts