Every documentation page is available as Markdown by replacing the trailing slash of its URL with ".md" (e.g. "/guides/introduction/" becomes "/guides/introduction.md") or by requesting it with the "Accept: text/markdown" header. An MCP server with tools to search and read this documentation is available at "/mcp". We also provide a machine-readable index of all Markdown resources at /llms.txt.

Coding agents

If you are using AI to generate Valibot schemas, our documentation provides several resources to help coding agents better understand the library. This page gives an overview of the available resources and how to use them.

Agent skill

Our SKILL.md contains specialized instructions for AI agents to write, migrate, and optimize Valibot schemas according to the latest best practices. You can install it by running the following command in your terminal.

npx skills add open-circle/agent-skills --skill valibot

The skill is also published at /.well-known/agent-skills/valibot/SKILL.md and discoverable via /.well-known/agent-skills/index.json.

MCP server

Our MCP server (Model Context Protocol) provides tools to search and read this documentation. Instead of crawling the website, coding agents can look up the right schema, method, or action in a single tool call. The server is available at https://valibot.dev/mcp and provides the following tools.

  • search_docs searches the guides and API reference and returns the most relevant pages
  • get_doc reads a documentation page and returns its full content as Markdown
  • list_docs lists all documentation pages grouped by area and category

The server is free to use, requires no authentication, and its metadata is published at /.well-known/mcp/server-card.json. To add it to Claude Code, run the following command.

claude mcp add --transport http valibot https://valibot.dev/mcp

For Cursor and other tools that use a JSON configuration file, add the following entry.

{
  "mcpServers": {
    "valibot": {
      "url": "https://valibot.dev/mcp"
    }
  }
}

LLMs.txt

An LLMs.txt file is a plain text file that provides instructions or metadata for large language models (LLMs). It often specifies how the LLMs should process or interact with content. It is similar to a robots.txt file, but is tailored for AI models.

We provide several LLMs.txt routes. Use the route that works best with your AI tool.

  • llms.txt contains a table of contents with links to Markdown files
  • llms-full.txt contains the Markdown content of the entire docs
  • llms-guides.txt contains the Markdown content of the guides
  • llms-api.txt contains the Markdown content of the API reference

Markdown for agents

We provide a Markdown version of every documentation page. You can access it by replacing the trailing slash (/) in the URL with .md. For example, /guides/installation/ becomes /guides/installation.md.

Alternatively, our server supports content negotiation. If a request to a documentation page contains the Accept: text/markdown header, the server responds with the Markdown version of the page instead of HTML.

Contributors

Thanks to all the contributors who helped make this page better!

  • GitHub profile picture of @fabian-hiller
  • GitHub profile picture of @flySewa

Partners

Thanks to our partners who support the project ideally and financially.

Sponsors

Thanks to our GitHub sponsors who support the project financially.

  • GitHub profile picture of @vasilii-kovalev
  • GitHub profile picture of @UpwayShop
  • GitHub profile picture of @ruiaraujo012
  • GitHub profile picture of @hyunbinseo
  • GitHub profile picture of @nickytonline
  • GitHub profile picture of @kibertoad
  • GitHub profile picture of @caegdeveloper
  • GitHub profile picture of @Thanaen
  • GitHub profile picture of @bmoyroud
  • GitHub profile picture of @ysknsid25
  • GitHub profile picture of @dslatkin