Ultra high quality Node.js library for managing:

• sitemap.xml fetching and parsing (supports http, https, file://)
• llms.txt generation, validation, and auto-correction (Markdown)
• Parsing llms.txt to structured JSON (from file or remote URL)

Designed for robustness, strict validation, and enterprise integration — using only native Node.js modules.

🚀 Features

• ✅ Robust sitemap.xml loading via HTTP(S) or local file system
• ✅ Smart HTTP ↔ HTTPS fallback logic
• ✅ Auto-generates clean llms.txt from structured data
• ✅ Validates and auto-corrects invalid Markdown input
• ✅ Parses llms.txt back to JSON format
• ✅ Built-in title extraction and priority filtering from sitemap
• ✅ Zero dependencies (uses native Node.js modules only)

📦 Installation

🧱 Basic Usage

import LLMSManager from "llms-manager";

(async () => {
  try {
    // 1. Create instance with sitemap URL
    const manager = new LLMSManager({
      sitemapUrl: "https://example.com/sitemap-pages.xml",
    });

    // 2. Load and parse the sitemap
    await manager.loadSitemap();
    console.log("Sitemap URLs:", manager.getSitemapEntries());

    // 3. Set title and description for llms.txt
    manager.setMetadata("My Site", "The best site ever");

    // 4. Automatically generate core content from sitemap (threshold 0.4)
    const coreContent = manager.autoGenerateCoreContent(0.4);
    manager.addCoreContent(coreContent);

    // 5. Add optional content manually
    manager.addOptionalContent([
      { title: "Extra Link", url: "https://example.com/extra" },
    ]);

    // 6. Generate llms.txt markdown (including validation & correction)
    const llmsText = manager.generateLLMSTxt();
    console.log("Generated llms.txt content:\n", llmsText);

    // 7. Save file
    await manager.saveToFile("./llms.txt");
    console.log("llms.txt saved successfully.");
  } catch (e) {
    console.error("Error:", e);
  }
})();

🧪 API Documentation

Constructor

new LLMSManager({ sitemapUrl: string });

• sitemapUrl: Required. Must begin with http://, https://, or file://.

Instance Methods

loadSitemap(): Promise<void>

Fetch and parse the sitemap to internal entries.

getSitemapEntries(): Array<{ loc: string, priority: number }>

Returns all parsed entries from the sitemap.

setMetadata(title: string, description: string): void

Set the title and description block for the generated llms.txt.

addCoreContent(items: Array<{ title: string, url: string }>): void

Add core content entries.

addOptionalContent(items: Array<{ title: string, url: string }>): void

Add optional content entries.

generateLLMSTxt(): string

Returns the final llms.txt markdown content as a string.

saveToFile(filepath: string): Promise<void>

Saves the generated llms.txt to a file.

autoGenerateCoreContent(threshold: number = 0.5): Array<{ title: string, url: string }>

Filters sitemap entries by priority and auto-generates a title from the URL.

Static Methods

LLMSManager.parseLLMSTxt(content: string): ParsedData

Parses llms.txt markdown into structured JSON:

{
  title: string;
  description: string;
  coreContent: Array<{ title: string, url: string }>;
  optionalContent: Array<{ title: string, url: string }>;
}

LLMSManager.parseLLMSTxtFromFile(filepath: string): Promise<ParsedData>

Loads and parses a local llms.txt file.

LLMSManager.parseLLMSTxtFromUrl(url: string): Promise<ParsedData>

Loads and parses a remote llms.txt file via HTTP/HTTPS.

📘 Example llms.txt Output

# My Project
> A curated list of intelligent content.

## Core Content
- [Home](https://example.com/)
- [Blog](https://example.com/blog)

## Optional
- [About](https://example.com/about)

💡 Tip

Use autoGenerateCoreContent() after loading the sitemap to quickly fill out relevant content based on priority.

🧩 Requirements

• Node.js v18+ (uses fs/promises and URL API)
• No external dependencies

🛠 Native Modules Used

• https, http
• fs/promises, fs
• url

🙋‍♂️ Contributing

Pull requests welcome! Please ensure your code is:

• Fully documented
• Tested
• Dependency-free

🔒 License

MIT — © 2025 Marius Heinrich