# Preferly

The Preferly MCP server for Claude Code checks whether AI crawlers can access a site and read its public business details. It also inspects llms.txt and structured business data.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI
- Author: preferly-ai
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/preferly-ai/preferly-mcp
- Updated: 2026-11-14
- Security check: automated, 2026-11-17 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/preferly

## What it does

Preferly checks a public website and returns its findings to the AI client. Its tools inspect crawler rules in `robots.txt`, the presence and format of `llms.txt`, structured business data, and whether a site's name, contact details, hours, and prices can be read without running scripts. Another check asks 3 AI models 3 typical customer questions and reports discrepancies with the website.

The server can also produce `robots.txt` lines with chosen access settings for search and training crawlers. Each tool is read-only and handles one public website per call. Preferly runs remotely over HTTP; no account or API key is needed.

## Requirements
- Remote server: `https://www.preferly.ai/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http preferly https://www.preferly.ai/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add preferly --url https://www.preferly.ai/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "preferly": {
            "url": "https://www.preferly.ai/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "preferly": {
            "type": "http",
            "url": "https://www.preferly.ai/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http preferly https://www.preferly.ai/mcp
```

## When to use it
- check whether `robots.txt` allows AI crawlers
- find format issues in `llms.txt`
- inspect business fields in a site's structured data
- check if hours and prices are readable without scripts
- compare AI model answers with a business website

## Example prompts
- Check whether AI crawlers can access `example.com`.
- Inspect the site's `llms.txt` and business schema for issues.
- Compare AI model answers about this business with `example.com`.

## Similar
- [SEO Web Analysis MCP — Site Audit & Tech Stack](https://affpapa.org/en/nevibecode/mcp/seo-web-analysis-mcp-site-audit-tech-stack.md) — MCP server · 9357 installs
- [Keyword Research — Google Suggest, Intent & Long-Tail](https://affpapa.org/en/nevibecode/mcp/keyword-research-google-suggest-intent-long-tail.md) — MCP server · 6583 installs
- [DomainKits MCP - the domain name intelligence](https://affpapa.org/en/nevibecode/mcp/domainkits-mcp-the-domain-name-intelligence.md) — MCP server · 5388 installs
- [Kakao Maps](https://affpapa.org/en/nevibecode/mcp/kakao-maps.md) — MCP server · 4882 installs
- [Earlywire — Marketing & Growth Intelligence](https://affpapa.org/en/nevibecode/mcp/earlywire-marketing-growth-intelligence.md) — MCP server · 4759 installs
- [Web Auditor](https://affpapa.org/en/nevibecode/mcp/web-auditor.md) — MCP server · 3818 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
