seo-insights

This MCP server exposes a set of SEO tooling capabilities to AI assistants via the Model Context Protocol. It connects to SEO APIs to provide backlinks analysis, keyword generation and evaluation, and traffic insights for websites. The CLI commands and MCP tools allow you to analyze a domain's backlinks, generate keyword ideas for a seed keyword, assess keyword difficulty and SERP details, and examine website traffic and top performing content. The server supports a stdio transport for local usage and a streamable HTTP transport for web-based clients; you can interact with the server either through the CLI or by configuring an HTTP RESTful MCP interface for integration with your frontend or agents.

Prerequisites: Install Node.js (version 18.x or newer) and Git.

Step 1: Clone the repository

git clone https://github.com/mrgoonie/seo-insights-mcp-server.git
cd seo-insights-mcp-server

Step 2: Install dependencies

npm install

Step 3: Run the development server (stdio transport by default)

npm run dev:server

Optional: Run the HTTP transport for remote usage

npm run dev:server:http

Notes: Ensure any required API keys (e.g., CAPSOLVER for captcha resolution) are provided via environment variables as described in the Environment Variables section of the README or your deployment configuration.

Environment and configuration tips:

• For HTTP transport, set MCP_HTTP_HOST, MCP_HTTP_PORT, and MCP_HTTP_PATH to configure the server endpoint.
• The MCP server expects a Node.js-based runtime when using the stdio transport; ensure the path to the built dist/index.js is correct in your mcp_config.
• If you plan to deploy remotely, remember to secure the HTTP transport and consider implementing authorization headers for web clients as indicated in the features notes.
• The project relies on external SEO APIs; manage API keys and request quotas according to the providers’ terms.
• If you encounter issues with the MCP Inspector or hot-reloading, ensure your development environment matches the project’s TypeScript and tooling setup (ESLint/Prettier/TypeScript) as described in the repository.