Skip to content
MCP server

llms.txt

Generate an llms.txt from Markdown or a sitemap and check an existing file

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, self-hosted
Runs
Locally
Version
0.3.0
Listing updated
2 days ago

About llms.txt

llms.txt is an open-source MCP server for ai models and llm tooling. It runs locally on your machine and is started by your AI client.

Everything stays local: the server reads files you name and returns text, never writing a file and never fetching a URL. Two generators produce the output. One scans a directory for .md, .mdx and .markdown files, takes titles from frontmatter, the first H1 or the file name, groups pages by top-level folder and lists anything skipped. The other builds one link per entry in a local sitemap.xml, grouped by first path segment, with titles derived from URLs because pages are not fetched.

The checker validates an llms.txt against the convention and reports errors, warnings and info with line numbers: a single H1 first, a blockquote summary, H2 sections of link lines, no duplicates and a modest size. Sitemap-derived files have empty notes you fill in yourself.

It runs over stdio from a pinned Docker image with a read-only mount, and the package is not yet on npmjs.com. It is one of ten small servers in dev-mcp-servers, MIT licensed, with no telemetry.

Install

Run in a terminal

claude mcp add mcp-llms-txt -- docker run -i --rm ghcr.io/basitalisandhu/mcp-llms-txt:0.3.0

Add to claude_desktop_config.json

{
  "mcpServers": {
    "mcp-llms-txt": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/basitalisandhu/mcp-llms-txt:0.3.0"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "mcp-llms-txt": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/basitalisandhu/mcp-llms-txt:0.3.0"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "mcp-llms-txt": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/basitalisandhu/mcp-llms-txt:0.3.0"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "mcp-llms-txt": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/basitalisandhu/mcp-llms-txt:0.3.0"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "mcp-llms-txt": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/basitalisandhu/mcp-llms-txt:0.3.0"
      ]
    }
  }
}

Tools this server provides (3)

Taken from the project documentation; the running server may differ.

generate_llms_txt
Scan a Markdown directory and return llms.txt text with entries and skipped files.
generate_from_sitemap
Build llms.txt links from a local sitemap.xml.
check_llms_txt
Check an llms.txt against the convention with line-numbered findings.

What it does

Markdown directory scan
Builds entries from up to 1000 Markdown files with titles and notes.
Sitemap conversion
Turns a local sitemap.xml into grouped llms.txt links.
Convention checker
Reports line-numbered errors and warnings on an existing file.
Local only
No network access; reads only the named files.
Docker distribution
Runs from a pinned image with a read-only mount.

Good for

  • Create a first llms.txt for a docs folder
  • Convert a sitemap into llms.txt entries
  • Lint an llms.txt before publishing
  • Find which Markdown files were skipped

Pricing

Open source

MIT licensed; free

Prices change. Check basitalisandhu.github.io before you buy.

llms.txt FAQ

Does it fetch my pages?

No. It never fetches URLs; sitemap titles come from the URL paths.

Does it write the file?

No. It returns text for you to review and save.

Can I use npx?

Not yet; the README says the package is not on npmjs.com, so use the Docker image or GitHub Packages.

What are the size limits?

Up to 1000 Markdown files of 1 MB each, or one sitemap up to 2 MB.

Is llms.txt free?

Yes. llms.txt is open source, so you can use and self-host it for free.

How do I add llms.txt to Claude Code?

Run this command in your terminal: claude mcp add mcp-llms-txt -- docker run -i --rm ghcr.io/basitalisandhu/mcp-llms-txt:0.3.0

What are the best alternatives to llms.txt?

The closest alternatives to llms.txt are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to llms.txt

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools