# mcp-typesense-server

> MCP Server

This repository provides a Model Context Protocol (MCP) server, enabling AI models to access Typesense search capabilities for discovering, searching, and analyzing data within Typesense collections. This repository provides a

## Overview

- **Category:** AI
- **Language:** TypeScript
- **Stars:** 19
- **Forks:** 0
- **Owner:** suhail-ak-s
- **GitHub:** https://github.com/suhail-ak-s/mcp-typesense-server
- **Created:** 2025-02-27T15:44:58+00:00
- **Updated:** 2025-03-19T14:49:25+00:00
- **Source:** https://model-context-protocol.com/servers/mcp-typesense-server-ai-model-search

## Setup

## Setup

1.  **Via npm:**
    -   Global installation:
        ```bash
        npm install -g typesense-mcp-server
        ```
    -   Local installation:
        ```bash
        npm install typesense-mcp-server
        ```
2.  **Via mcp-get:**
    ```bash
    npx @michaellatman/mcp-get@latest install typesense-mcp-server
    ```
3.  **Development:**
    -   Install dependencies:
        ```bash
        npm install
        ```
    -   Build the server:
        ```bash
        npm run build
        ```
    -   For development with auto-rebuild:
        ```bash
        npm run watch
        ```
4.  **Installation for Development Using Claude Desktop:**
    -   Add the server config to `claude_desktop_config.json` (On MacOS: `~/Library/Application Support/Claude/claude_desktop_config.json`, On Windows: `%APPDATA%/Claude/claude_desktop_config.json`) with the provided JSON configuration, replacing placeholders with your Typesense host and API key.
5.  **Usage with Claude Desktop:**
    -   Add the provided configuration to the "mcpServers" section of your `claude_desktop_config.json`, replacing placeholders with your Typesense host and API key.

## Tools

## Available Tools

1.  **Resources:** List and access collections via `typesense://` URIs, with each collection having a name, description, and document count, and JSON mime type for schema access.
2.  **typesense_query:** Search for documents in Typesense collections with powerful filtering, taking query text, collection name, search fields, filters, sort options, and limit as input, and returning matching documents with relevance scores.
3.  **typesense_get_document:** Retrieve specific documents by ID from collections, taking collection name and document ID as input, and returning complete document data.
4.  **typesense_collection_stats:** Get statistics about a Typesense collection, taking collection name as input, and returning collection metadata, document count, and schema information.
5.  **analyze_collection:** Analyze collection structure and contents, taking collection name as input, and outputting insights about schema, data types, and statistics.
6.  **search_suggestions:** Get suggestions for effective search queries for a collection, taking collection name as input, and outputting recommended search strategies based on collection schema.