Connect LLMs to Typesense Data
Typesense MCP Server gives AI assistants search access to Typesense collections - querying, fetching documents, and stats.
Why it matters
Empower AI models with Typesense search capabilities. Enable LLMs to seamlessly find, search, and analyze data stored within your Typesense collections.
Outcomes
What it gets done
Integrate LLMs with Typesense for data retrieval and analysis.
Enable full-text search, filtering, and sorting of Typesense collections.
Provide access to collection statistics, metadata, and schemas.
Facilitate document retrieval by ID.
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/vb-typesense | bash Capabilities
Tools your agent gets
Search documents in Typesense collections with filtering, query text, and customizable parameters
Retrieve specific documents by ID from collections, returning complete document data
Retrieve statistics about a Typesense collection including metadata, document count, and schema
List all available collections with their schemas for discovery and routing
Overview
Typesense MCP Server
An MCP server giving an AI assistant direct access to Typesense: discover collections, run full-text search with filters and sorting, fetch documents by ID, and pull collection stats. Use it when an assistant needs to search or analyze data already indexed in Typesense - it's a read/query layer, not an admin tool.
What it does
Typesense MCP Server is a Model Context Protocol implementation that gives an AI assistant direct access to Typesense search capabilities. It lets an LLM discover what collections exist, run full-text searches with filtering, sorting, and pagination, retrieve a specific document by ID, and pull statistics and schema information about a collection - all without leaving the chat interface. Collections are also exposed as first-class MCP resources at typesense://collections/<collection> URIs, returning JSON schema information (field names and data types) plus sample documents so an assistant can inspect a collection's shape before querying it, and the server ships resource templates for constructing search queries and viewing collection details.
When to use - and when NOT to
Use this connector when an assistant needs to search or analyze data already indexed in Typesense: answering questions grounded in a product catalog, document store, or any searchable dataset, or helping a user construct effective search queries against a collection whose schema they don't know by heart. It is not useful without an existing Typesense deployment to connect to, and it does not create or index collections - it is a read/query layer, not a Typesense admin tool.
Capabilities
typesense_list_collections: lists all available collections with schema and field definitions (searchable, facetable, numeric fields), supporting aninclude_fieldsparameter. It enables zero-config discovery - the LLM can enumerate collections at runtime and pick the right one(s) before searching, which matters when collections vary by environment, tenant, or version.typesense_query: full-text searches a collection with query text, search fields, filters, sort options, and limit, returning matches with relevance scores.typesense_get_document: retrieves a specific document by collection name and document ID.typesense_collection_stats: returns collection metadata, document count, and schema information.- Collections are also exposed as MCP resources via
typesense://URIs, and promptsanalyze_collectionandsearch_suggestionshelp interpret schema and suggest search strategies.
How to install
Install via npm (globally, or locally within a project):
npm install -g typesense-mcp-server
Or install via mcp-get:
npx @michaellatman/mcp-get@latest install typesense-mcp-server
For Claude Desktop, add an mcpServers entry in claude_desktop_config.json with command: npx, args ["-y", "typesense-mcp-server", "--host", "your-typesense-host", "--port", "8108", "--protocol", "http", "--api-key", "your-api-key"]. For debugging, use the MCP Inspector via npm run inspector; server logs are written to /tmp/typesense-mcp.log.
Who it's for
Teams running Typesense who want their AI assistant to search, retrieve, and reason over indexed data directly, or to help users build better search queries against collections whose structure varies by environment or tenant. The server is licensed under the MIT License.
Source README
Typesense MCP Server
A Model Context Protocol (MCP) server implementation that provides AI models with access to Typesense search capabilities. This server enables LLMs to discover, search, and analyze data stored in Typesense collections.
Demo
Features
Resources
- List and access collections via
typesense://URIs - Each collection has a name, description, and document count
- JSON mime type for schema access
Tools
typesense_query
- Search for documents in Typesense collections with powerful filtering
- Input: Query text, collection name, search fields, filters, sort options, limit
- Returns matching documents with relevance scores
typesense_get_document
- Retrieve specific documents by ID from collections
- Input: Collection name, document ID
- Returns complete document data
typesense_collection_stats
- Get statistics about a Typesense collection
- Input: Collection name
- Returns collection metadata, document count, and schema information
Prompts
analyze_collection
- Analyze collection structure and contents
- Input: Collection name
- Output: Insights about schema, data types, and statistics
search_suggestions
- Get suggestions for effective search queries for a collection
- Input: Collection name
- Output: Recommended search strategies based on collection schema
Installation
Via npm
# Global installation
npm install -g typesense-mcp-server
# Local installation
npm install typesense-mcp-server
Via mcp-get
npx @michaellatman/mcp-get@latest install typesense-mcp-server
Development
Install dependencies:
npm install
Build the server:
npm run build
For development with auto-rebuild:
npm run watch
Installation for Development
Using Claude Desktop
To use with Claude Desktop, add the server config:
On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
On Windows: %APPDATA%/Claude/claude_desktop_config.json
{
"mcpServers": {
"typesense": {
"command": "node",
"args": [
"~/typesense-mcp-server/dist/index.js",
"--host", "your-typesense-host",
"--port", "8108",
"--protocol", "http",
"--api-key", "your-api-key"
]
},
}
}
Debugging
Since MCP servers communicate over stdio, debugging can be challenging. We recommend using the MCP Inspector, which is available as a package script:
npm run inspector
The Inspector will provide a URL to access debugging tools in your browser.
Components
Resources
The server provides information about Typesense collections:
- Collection Schemas (
typesense://collections/<collection>)- JSON schema information for each collection
- Includes field names and data types
- Sample documents for understanding data structure
Tools
The server provides the following tools for interacting with Typesense:
typesense_list_collections - List all available collections with their schemas
- Enables zero-conf discovery and routing
- LLM can enumerate collections at runtime and pick the right one(s) before searching
- Useful when collections vary by environment, tenant, or version
- Returns field definitions for schema inference (searchable, facetable, numeric fields)
- Supports
include_fieldsparameter to control detail level
typesense_query - Search for documents in a collection
- Full-text search with customizable parameters
- Supports filtering, sorting, and pagination
typesense_get_document - Retrieve a specific document by ID
typesense_collection_stats - Get detailed statistics about a specific collection
Resource Templates
The server provides templates for:
- typesense_search - Template for constructing Typesense search queries
- typesense_collection - Template for viewing Typesense collection details
Usage with Claude Desktop
To use this server with the Claude Desktop app, add the following configuration to the "mcpServers" section of your claude_desktop_config.json:
{
"mcpServers": {
"typesense": {
"command": "npx",
"args": [
"-y",
"typesense-mcp-server",
"--host", "your-typesense-host",
"--port", "8108",
"--protocol", "http",
"--api-key", "your-api-key"
]
}
}
}
Logging
The server logs information to a file located at:
/tmp/typesense-mcp.log
This log contains detailed information about server operations, requests, and any errors that occur.
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.

