Llms.txt Documentation

🚀 mcpdoc Documentation Tool

mcpdoc is a tool for loading and querying documentation. It supports multiple ways to configure and manage document sources, enabling users to quickly access and search for relevant documentation.

🚀 Quick Start

1. Start the Documentation Server

Start the documentation server with the following command:

mcpdoc --urls LangGraph:https://langchain-ai.github.io/langgraph/llms.txt

This command will load and serve the LangGraph Python documentation.

2. Query the Documentation

After starting the server, you can query the documentation as follows:

/mcp query "Please list all supported models"

📦 Installation

Install using pip:

pip install mcpdoc

📚 Documentation

The documentation server supports various configuration methods, allowing you to flexibly add and manage document sources.

1. Use a YAML Configuration File

Create a sample_config.yaml file:

# Sample configuration for mcpdoc server
# Each entry must have a llms_txt URL and can optionally include a name
- name: LangGraph Python
  llms_txt: https://langchain-ai.github.io/langgraph/llms.txt

Specify the YAML configuration file when starting the server:

mcpdoc --yaml sample_config.yaml

2. Use a JSON Configuration File

Create a sample_config.json file:

[
  {
    "name": "LangGraph Python",
    "llms_txt": "https://langchain-ai.github.io/langgraph/llms.txt"
  }
]

Specify the JSON configuration file when starting the server:

mcpdoc --json sample_config.json

3. Specify URLs Directly

You can directly specify the URLs of document sources using the --urls parameter in the format name:url. For example:

mcpdoc --urls LangGraph:https://langchain-ai.github.io/langgraph/llms.txt --urls LangChain:https://python.langchain.com/llms.txt

4. Combine Usage

You can combine the use of YAML, JSON configuration files, and the --urls parameter:

mcpdoc --yaml sample_config.yaml --json sample_config.json --urls LangGraph:https://langchain-ai.github.io/langgraph/llms.txt --urls LangChain:https://python.langchain.com/llms.txt

🔧 Additional Options

• --follow-redirects: Whether to follow HTTP redirects (default is False).
• --timeout SECONDS: HTTP request timeout in seconds (default is 10.0).

Example:

mcpdoc --yaml sample_config.yaml --follow-redirects --timeout 15

💻 Usage Examples

Basic Usage

from mcpdoc.main import create_server

# Create a server instance and specify document sources
server = create_server(
    [
        {
            "name": "LangGraph Python",
            "llms_txt": "https://langchain-ai.github.io/langgraph/llms.txt",
        },
        # You can add multiple document sources
        # {
        #     "name": "Another Documentation",
        #     "llms_txt": "https://example.com/llms.txt",
        # },
    ],
    follow_redirects=True,
    timeout=15.0,
)

# Start the server
server.run(transport="stdio")