Perplexity Search MCP

# Perplexity Search MCP This tool wraps the search functionality of [Perplexity AI](https://www.perplexity.ai/) into a standardized tool using the [Model Context Protocol (MCP)](https://modelcontextprotocol.io/docs/getting-started/intro). It allows you to perform advanced search queries with filtering and customization options. This project is built with TypeScript and runs on the [Bun](https://bun.sh/) runtime. ## ✨ Features - Provides the `perplexity-search` tool via MCP. - Supports various parameters for Perplexity search: - `query`: A single search query or an array of queries for multi-query search. - `max_results`: Specifies the maximum number of search results to return (maximum 20). - `country`: Filters search results by geographic location (e.g., 'US', 'GB', 'DE'). - `max_tokens_per_page`: Controls the maximum number of tokens retrieved from each webpage during search processing. Higher values provide more comprehensive content extraction but may increase processing time. - `search_domain_filter`: A list of domains/URLs to restrict search results. For example: `["science.org", "pnas.org", "cell.com"]` - Communicates with the host environment via standard input/output (Stdio). ## 🚀 Getting Started ### 1. Prerequisites - Ensure you have [Bun](https://bun.sh/docs/installation) installed. - Obtain an API key from Perplexity AI. ### 2. Installation After cloning the project, run the following command in the root directory to install the required dependencies: ```bash bun install ``` ### 3. Configuration Set the `PERPLEXITY_API_KEY` environment variable with your Perplexity API key. ### 4. Running the Server (Development) To run the server directly from the TypeScript source for development, execute the following command: ```bash bun run index.ts ``` The server will start and listen for MCP messages on standard I/O. ### 5. Building the Project You can build the project into a single, optimized JavaScript file: ```bash bun run build ``` This compiles `index.ts` and places the output in `dist/perplexity-search-mcp.js`. After building, you can run the compiled file directly: ```bash bun dist/perplexity-search-mcp.js ``` or ```bash node dist/perplexity-search-mcp.js ``` ## 🛠️ Tool Definition ### `perplexity-search` **Description**: Retrieve ranked search results from Perplexity’s continuously refreshed index, with advanced filtering and customization options. **Input Parameters**: | Parameter | Type | Description | Default | | --------------------- | ------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------- | | `query` | `string` or `string[]` | The search query or queries to execute. Can be a single query or a list of queries for multi-query search. | Required | | `max_results` | `number` | The maximum number of search results to return (must be a positive integer no greater than 20). | `10` | | `country` | `string` | Country code to filter search results by geographic location (e.g., 'US', 'GB', 'DE'). | Optional | | `max_tokens_per_page` | `number` | Maximum number of tokens to retrieve from each webpage. Higher values yield more comprehensive content extraction but may increase processing time. | `2048` | | `search_domain_filter`| `string[]` | An optional list of domains/URLs to which search results should be limited (e.g., `["science.org", "pnas.org", "cell.com"]`). | Optional | **Output**: The tool returns a JSON string containing an array of search results from the Perplexity API.