~/docs

Use the docs with AI

Read these docs from coding agents and AI assistants through the MCP server, llms.txt or plain Markdown.

↗ .mdmcp

Every page of these docs is available as plain Markdown, so AI assistants and coding agents can read the same content you do: through an MCP server, through llms.txt, or page by page.

MCP server

The shellz docs MCP server is at:

https://shellz.dev/mcp

It uses the Streamable HTTP transport and needs no authentication. It's read-only and only serves these docs.

ToolWhat it does
search_docsFull-text search. Takes a query and an optional limit, and returns matching pages and sections with their paths.
list_docsThe table of contents: every page with its title, description and Markdown URL.
read_docThe full Markdown of one page. Takes a path such as /docs/install/kubernetes, or a full URL.

Every page is also exposed as an MCP resource (https://shellz.dev/docs/<page>.md, type text/markdown).

Add it to your client

// ~/.cursor/mcp.json, or .cursor/mcp.json in your project
{
  "mcpServers": {
    "shellz-docs": { "url": "https://shellz.dev/mcp" }
  }
}

Other clients that support remote MCP servers over HTTP work the same way: add https://shellz.dev/mcp as the server URL.

Then ask, for example: "Using the shellz docs, add shellz to our Kubernetes deployment" or "How does shellz's Perfect Secrecy use passkeys?"

Discovery

Clients and registries that support MCP Server Cards can find the server without configuration:

The card is also served at /.well-known/mcp.json and /.well-known/mcp/server-card.json for clients that use those earlier draft locations.

llms.txt

For tools that read llms.txt:

Markdown for any page

  • Add .md to any docs URL: https://shellz.dev/docs/getting-started.md.

  • Or request a docs URL with the header Accept: text/markdown:

    curl -H 'Accept: text/markdown' https://shellz.dev/docs/install/linux
  • On each page, copy markdown copies the page and .md opens it.