home / mcp / doc bot mcp server

Doc Bot MCP Server

Provides live, searchable project docs and API references to AI assistants via MCP for enhanced coding guidance.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "afterxleep-doc-bot": {
      "command": "npx",
      "args": [
        "@afterxleep/doc-bot@latest"
      ]
    }
  }
}

You can empower AI assistants with your project’s knowledge by running doc-bot as an MCP server. It indexes your docs, surfaces relevant guidance contextually, and updates your documentation as patterns evolve, so your AI helpers stay accurate and helpful.

How to use

Install and run doc-bot as an MCP server so your AI assistants can access your project documentation. The server works with any MCP-compatible assistant and focuses on smart search, contextual guidance, and live doc updates. You interact with it by configuring your MCP client to connect to the local or remote server, then asking your assistant questions like “What are the coding standards for this project?” or “Show me the auth flow.” The integration lets the AI surface targeted docs and API references without overloading the prompt.

How to install

Prerequisites: Node.js and npm (or a compatible Node runtime) must be installed on your machine.

1. Add doc-bot to your MCP client configuration using the stdio (local) setup shown below.

{ 
  "mcpServers": {
    "doc-bot": {
      "command": "npx",
      "args": ["@afterxleep/doc-bot@latest"]
    }
  }
}

Configuration and usage notes

Create a project documentation folder named doc-bot at the root of your project and add markdown files to seed your knowledge base. For example, you can include coding standards, API patterns, testing guides, and architecture notes.

Example project structure you can start with:

your-project/
├── doc-bot/
│   ├── coding-standards.md
│   ├── api-patterns.md
│   ├── testing-guide.md
│   └── architecture.md
├── src/
└── package.json

Doc formats and frontmatter

Document files use frontmatter to describe the doc’s metadata, including title, description, keywords, and file patterns that determine when they apply.

---
title: "React Component Guidelines"
description: "Standards for building React components"
keywords: ["react", "components", "frontend", "jsx"]
---

# React Component Guidelines

- Use functional components with hooks
- Follow PascalCase naming
- Keep components under 200 lines
- Write tests for all components

Agent-driven updates and docset highlights

doc-bot enables agents to update the documentation when new patterns or changes are discovered. You can also leverage Docsets to access official API references for frameworks and libraries from installed documentation databases.

How to test it

Once the server is running, ask your AI assistant to fetch relevant docs or to read a specific file’s documentation. For example, request the coding standards or a particular API reference to verify compatibility with your current task.

Available tools

doc_bot

Get documentation guidance for the current project task

search_documentation

Search all project and API documentation for relevant results

get_file_docs

Retrieve documentation for a specific file or path

read_specific_document

Open a full documentation document by name or title

get_document_index

List all available documentation in the project store

create_or_update_rule

Add or update documentation entries when new patterns emerge

refresh_documentation

Reload documentation from disk to ensure the latest content is used

explore_api

Explore API documentation and related classes/methods

add_docset

Install a new API docset for official references

remove_docset

Remove an installed API docset

list_docsets

List all installed API docsets