Improved developer docs for Agents

TLDR: We've made three improvements to llms.txt in Fern Docs:

1. API Reference pages: llms.txt now includes full OpenAPI specifications and code samples for all enabled languages from every API endpoint page.
2. Automatic Markdown serving: When Fern detects an LLM bot, it automatically serves Markdown content instead of HTML, reducing token consumption by 90%+.‍
3. Analytics in the Fern Dashboard: Track which AI providers (Claude, ChatGPT, Cursor, etc.) are accessing your documentation and see traffic breakdowns between bots and humans.

API Reference pages now support llms.txt

Every API endpoint page on your Fern Docs site now includes its full OpenAPI specification and code samples in all the languages you’ve enabled in the llms.txt format. This makes it easy for AI tools to understand your API structure and helps developers implement your endpoints correctly.

Example from Cash App (Create speech endpoint): https://developers.cash.app/cash-app-afterpay/api-reference/reference/disputes/retrieve-dispute/llms.txt

"Fern helped us future-proof our docs. With llms.txt, AI agents can crawl and understand every API endpoint without my team doing any duplicative work. What used to be a nice-to-have is the new standard for modern developer experiences."

- Erin Nielsen, Senior Technical Writer at Cash App

Reduce token costs by serving Markdown to bots

When Fern detects an LLM bot accessing your documentation, it automatically serves Markdown content from llms.txt files instead of HTML. This reduces token consumption by over 90%, saving costs for both you and the bots crawling your site.

This is especially useful for:

• Bot crawlers: Automatically reduce token usage when AI tools index your documentation
• Developer workflows: Your customers can copy API Reference pages as Markdown to paste into AI assistants
• Usage insights: Monitor how AI tools are consuming your documentation

Track llms.txt usage with analytics

You can now see exactly how AI agents interact with your documentation in the Fern Dashboard. The new analytics include:

• Page-level visitor breakdown: See which pages are being accessed by AI bots versus human visitors
• LLM bot traffic by provider: See which AI tools (Claude, ChatGPT, Cursor, etc.) are accessing your docs

This visibility helps you identify which pages to optimize for AI consumption and understand how developers are using AI tools with your documentation.

Get started with Fern Docs today

When you use Fern Docs, your llms.txt files are automatically generated and maintained – no configuration required. Ready to make your documentation more agent-friendly?

‍Reach out to learn how Fern can help.

‍