The Future of Documentation is Rule-Driven

🔍

Rule-Based Linting

Lint documents against specified rules with customizable standards that reflect your organization's voice and structure.

🔧

Auto-Fix Capability

Automatically fix linting errors and ensure consistent documentation across your entire project.

🧩

Flexible Rule System

Create, share, and extend rules with a hierarchical system that respects rule priority and resolves conflicts.

🔌

MCP Server Integration

Seamless integration with AI tools through Model Context Protocol servers for enhanced capabilities.

🌐

Remote Rules

Load rules from local files or remote repositories to maintain consistent standards across teams.

⚡

Performance Optimization

Rule packing and caching for improved performance, making documentation linting efficient at scale.

Overview

AI-Lint is an experimental project exploring AI capabilities for document linting. It provides a flexible framework for defining and enforcing documentation standards through a rule-based system. The tool integrates with AI platforms like Cline and other AI coding tools via MCP (Model Context Protocol) servers.

bash

# Quick installation
git clone https://github.com/ToyB0x/ai-lint.git
cd ai-lint
pnpm install && pnpm build

ConfigurationUsageRules

json

{
  "mcpServers": {
    "@ai-lint/mcp": {
      "command": "node",
      "args": [
        "/PATH/TO/YOUR_DIR/ai-lint/apps/mcp/dist/index.js"
      ],
      "autoApprove": [],
      "disabled": false
    }
  }
}

markdown

You (AI assistants like Cline or Roo Code) must follow these important rules:

- Verify that the MCP Server @ai-lint/mcp is properly set up
- If not set up correctly, inform the user "The @ai-lint/mcp configuration is incorrect" and halt all subsequent processing
- If set up correctly, execute @ai-lint/mcp's load-remote-act-rule and apply the following rules