SEO MCP
A MCP (Model Control Protocol) SEO tool service based on Ahrefs data. Includes features such as backlink analysis, keyword research, traffic estimation, and more.
[ ](./README_CN.md)
Overview
This service provides an API to retrieve SEO data from Ahrefs. It handles the entire process, including solving the CAPTCHA, authentication, and data retrieval. The results are cached to improve performance and reduce API costs.
This MCP service is for educational purposes only. Please do not misuse it. This project is inspired by `@ `.
Features
- Backlink Analysis
- Get detailed backlink data for any domain
- View domain rating, anchor text, and link attributes
- Filter educational and government domains
- Keyword Research
- Generate keyword ideas from a seed keyword
- Get keyword difficulty score
- View search volume and trends
- Traffic Analysis
- Estimate website traffic
- View traffic history and trends
- Analyze popular pages and country distribution
- Track keyword rankings
- Performance Optimization
- Use CapSolver to automatically solve CAPTCHA
- Response caching
Installation
Prerequisites
- Python 3.10 or higher
- CapSolver account and API key ([register here](https://dashboard.capsolver.com/passport/register?inviteCode=1dTH7WQSfHD0))
Install from PyPI
Or use `uv`:
Manual Installation
- Clone the repository: ```bash git clone https://github.com/cnych/seo-mcp.git cd seo-mcp ```
- Install dependencies: ```bash pip install -e . # Or uv pip install -e . ```
- Set the CapSolver API key: ```bash export CAPSOLVER_API_KEY="your-capsolver-api-key" ```
Usage
Run the service
You can run the service in the following ways:
Use in Cursor IDE
In the Cursor settings, switch to the MCP tab, click the `+Add new global MCP server` button, and then input:
You can also create a `.cursor/mcp.json` file in the project root directory, with the same content.
API Reference
The service provides the following MCP tools:
`get_backlinks_list(domain: str)`
Get the backlinks of a domain.
**Parameters:**
- `domain` (string): The domain to analyze (e.g. "example.com")
**Returns:**
`keyword_generator(keyword: str, country: str = "us", search_engine: str = "Google")`
Generate keyword ideas.
**Parameters:**
- `keyword` (string): The seed keyword
- `country` (string): Country code (default: "us")
- `search_engine` (string): Search engine (default: "Google")
**Returns:**
`get_traffic(domain_or_url: str, country: str = "None", mode: str = "subdomains")`
Get the traffic estimation.
**Parameters:**
- `domain_or_url` (string): The domain or URL to analyze
- `country` (string): Country filter (default: "None")
- `mode` (string): Analysis mode ("subdomains" or "exact")
**Returns:**
`keyword_difficulty(keyword: str, country: str = "us")`
Get the keyword difficulty score.
**Parameters:**
- `keyword` (string): The keyword to analyze
- `country` (string): Country code (default: "us")
**Returns:**
Development
For development:
How it works
- The user sends a request through MCP
- The service uses CapSolver to solve the Cloudflare Turnstile CAPTCHA
- The service gets the authentication token from Ahrefs
- The service retrieves the requested SEO data
- The service processes and returns the formatted results
Troubleshooting
- **CapSolver API key error** Check the `CAPSOLVER_API_KEY` environment variable
- **Rate limiting** Reduce request frequency
- **No results** The domain may not be indexed by Ahrefs
- **Other issues** See [GitHub repository](https://github.com/cnych/seo-mcp)
License
MIT License - See LICENSE file