#Lynx for AI

LLMs are trained on public web data, which means they often lack awareness of the latest Lynx features and best practices. The Lynx team and community are developing a suite of tools to help coding agents understand and develop Lynx more effectively.

#Set Up Your Project with AGENTS.md

The AGENTS.md file is similar to README.md, but written for coding agents. It gives agents the project-specific context they need to work effectively in your Lynx codebase, with zero dependencies and immediate value.

Learn how to add a suitable AGENTS.md for your Lynx project in AGENTS.md for Lynx.

#Add Domain Expertise with Agent Skills

Agent Skills are reusable rule sets that give coding agents domain-specific knowledge about Lynx's dual-thread architecture, common patterns, and debugging techniques. Install the community collection with:

npx skills add lynx-community/skills

Read more in Agent Skills for Lynx.

#Connect Tools with MCP

Model Context Protocol is a standard for connecting coding agents to external tools.

• Lynx Docs MCP allows agents to fetch accurate guides and API references directly from the Lynx documentation site, ensuring answers stay correct and up to date.
• Lynx DevTool MCP lets coding agents control, operate and preview Lynx pages, just like a human does in DevTool.

#Using llms.txt

The Lynx website also supports the llms.txt standard, which is useful for feeding Lynx documentation into any LLM with web browsing. For every page, you can get the original Markdown by replacing the .html extension with .md. For example:

• https://lynxjs.org/llms.txt
• https://lynxjs.org/react/introduction.md