Configuration

This page covers every configuration option across the three main config objects.

ProxyConfig

Passed to withMarkdownMirror() in proxy.ts.

Option	Type	Default	Description
`routePrefix`	`string`	`"/md-mirror"`	URL prefix for the Markdown route handler. Must match the folder name in `app/`.
`excludePaths`	`string[]`	`[]`	Path prefixes to skip — requests to these paths pass through without interception.

Example

proxy.ts
import { withMarkdownMirror } from 'next-markdown-mirror/nextjs';

export const proxy = withMarkdownMirror({
  routePrefix: '/md-mirror',
  excludePaths: ['/api', '/admin', '/_next'],
});

RouteHandlerConfig

Passed to createMarkdownHandler() in your route handler. Extends MarkdownMirrorConfig with additional options.

Option	Type	Default	Description
`baseUrl`	`string`	required	The base URL of your site (e.g. `https://example.com`). Used to fetch pages internally.
`additionalHeaders`	`Record<string, string>`	`{}`	Extra headers to include in the internal fetch request (e.g. auth tokens).

Plus all options from MarkdownMirrorConfig below.

Example

app/md-mirror/[...path]/route.ts
import { createMarkdownHandler } from 'next-markdown-mirror/nextjs';

export const GET = createMarkdownHandler({
  baseUrl: process.env.NEXT_PUBLIC_SITE_URL!,
  contentSelectors: ['article', '.blog-post'],
  excludeSelectors: ['.sidebar', '.comments'],
  extractJsonLd: true,
  additionalHeaders: {
    'Cache-Control': 'public, max-age=3600',
  },
});

MarkdownMirrorConfig

The core conversion configuration. Used by RouteHandlerConfig and the HtmlToMarkdown class.

Option	Type	Default	Description
`contentSelectors`	`string[]`	`["main", "article", "[role=\"main\"]", "#content", "#main-content", ".main-content", ".post-content", ".entry-content", ".article-content"]`	Priority-ordered CSS selectors to find the main content. The first match is used; falls back to `<body>`.
`excludeSelectors`	`string[]`	`[]`	Additional CSS selectors for elements to remove from content. Added on top of the built-in exclusions.
`includeSelectors`	`string[]`	`[]`	CSS selectors for elements to preserve even if they would normally be excluded.
`extractJsonLd`	`boolean`	`true`	Extract JSON-LD structured data from `<script type="application/ld+json">` tags and include as YAML frontmatter.
`turndownRules`	`Record<string, TurndownService.Rule>`	`{}`	Custom Turndown conversion rules. See Custom Turndown Rules.
`turndownOptions`	`TurndownService.Options`	`{}`	Override default Turndown options (heading style, bullet list marker, etc.).
`baseUrl`	`string`	`undefined`	Base URL for resolving relative links and images.
`maxContentSize`	`number`	`1048576` (1 MB)	Maximum HTML size in bytes before rejecting. Throws an error if exceeded.
`contentSignal`	`ContentSignal`	`undefined`	Sets the `Content-Signal` response header. One of `"ai-train"`, `"search"`, or `"ai-input"`.
`tokenCounter`	`(text: string) => number`	built-in	Custom function to count tokens. The built-in counter estimates based on whitespace splitting.

Content selectors example

export const GET = createMarkdownHandler({
  baseUrl: process.env.NEXT_PUBLIC_SITE_URL!,
  // Only look for content in these containers
  contentSelectors: ['.docs-content', 'article', 'main'],
  // Remove sidebars and table of contents
  excludeSelectors: ['.sidebar', '.toc', '.breadcrumbs'],
  // Keep code blocks even if they're inside excluded areas
  includeSelectors: ['pre', 'code'],
});

Turndown options example

export const GET = createMarkdownHandler({
  baseUrl: process.env.NEXT_PUBLIC_SITE_URL!,
  turndownOptions: {
    headingStyle: 'atx',        // # Heading (default)
    bulletListMarker: '-',       // - item (default is *)
    codeBlockStyle: 'fenced',   // ```code``` (default)
    emDelimiter: '_',            // _emphasis_ (default is *)
  },
});

LlmsTxtConfig

Passed to createLlmsTxtHandler() and createLlmsFullTxtHandler().

Option	Type	Default	Description
`siteName`	`string`	required	Your site's name. Appears as the title in llms.txt.
`baseUrl`	`string`	required	Base URL of your site. Used to build absolute page URLs.
`description`	`string`	`undefined`	A description of your site shown below the title.
`pages`	`LlmsPage[] \| string`	required	Array of page objects, or a URL string pointing to a sitemap.xml.
`sections`	`Record<string, { title: string; description?: string }>`	`undefined`	Group pages into named sections. Keys match the `section` field on `LlmsPage` objects.

LlmsPage object

Field	Type	Required	Description
`url`	`string`	yes	Page path (e.g. `/about`).
`title`	`string`	yes	Page title shown in llms.txt.
`description`	`string`	no	Brief description of the page.
`section`	`string`	no	Section key this page belongs to. Must match a key in `sections`.

Example with sections

app/llms.txt/route.ts
import { createLlmsTxtHandler } from 'next-markdown-mirror/nextjs';

export const GET = createLlmsTxtHandler({
  siteName: 'My Docs',
  baseUrl: 'https://docs.example.com',
  description: 'Technical documentation for the Example API.',
  sections: {
    guides: { title: 'Guides', description: 'Step-by-step tutorials' },
    reference: { title: 'API Reference' },
  },
  pages: [
    { url: '/', title: 'Home', description: 'Welcome' },
    { url: '/quickstart', title: 'Quick Start', section: 'guides' },
    { url: '/authentication', title: 'Authentication', section: 'guides' },
    { url: '/endpoints', title: 'Endpoints', section: 'reference' },
    { url: '/errors', title: 'Error Codes', section: 'reference' },
  ],
});

See llms.txt Protocol for more details including sitemap support and the full-text variant.

ProxyConfig​

Example​

RouteHandlerConfig​

Example​

MarkdownMirrorConfig​

Content selectors example​

Turndown options example​

LlmsTxtConfig​

LlmsPage object​

Example with sections​

ProxyConfig

Example

RouteHandlerConfig

Example

MarkdownMirrorConfig

Content selectors example

Turndown options example

LlmsTxtConfig

LlmsPage object

Example with sections