MindsDB MySQL MCP Server
Connects AI clients to MindsDB via the MySQL protocol to execute SQL queries, manage databases, and perform semantic searches within knowledge bases. It enables automated workflows through job scheduling and provides seamless integration with external data sources.
README
MindsDB MySQL MCP Server
MCP server for MindsDB that connects via MySQL protocol (port 47335), providing reliable tools for Claude Code and Desktop.
Why MySQL over SSE/HTTP?
- MySQL protocol is stable and well-tested
- No async event loop issues (the "this event loop is already running" errors)
- Better error messages and debugging
- Works reliably with knowledge base queries
Installation
cd mcp/mindsdb-mysql
npm install
npm run build
Configuration
Environment Variables
| Variable | Default | Description |
|---|---|---|
MINDSDB_HOST |
127.0.0.1 |
MindsDB server host |
MINDSDB_PORT |
47335 |
MindsDB MySQL port |
MINDSDB_USER |
mindsdb |
Username (use your MindsDB Cloud email for cloud instances) |
MINDSDB_PASSWORD |
(empty) | Password (required for MindsDB Cloud) |
Claude Code (~/.claude.json)
{
"mcpServers": {
"mindsdb": {
"type": "stdio",
"command": "node",
"args": [
"/path/to/mindsdb-mysql/dist/index.js"
],
"env": {
"MINDSDB_HOST": "localhost",
"MINDSDB_PORT": "47335",
"MINDSDB_USER": "mindsdb",
"MINDSDB_PASSWORD": ""
}
}
}
}
For MindsDB Cloud, use your email as MINDSDB_USER and your password for MINDSDB_PASSWORD.
Claude Desktop (claude_desktop_config.json)
Same configuration, add to the mcpServers object.
Available Tools (11)
1. query - Execute SQL
Execute any MindsDB SQL query. Use this for custom queries not covered by other tools.
query(sql: string, database?: string): QueryResult
2. list_databases - List Data Sources
List all databases/data sources in MindsDB.
list_databases(): string[]
3. list_knowledge_bases - List KBs
List all knowledge bases in MindsDB.
list_knowledge_bases(): KnowledgeBase[]
4. describe - Describe Any Object
Describe a model, table, or knowledge base. Returns column information.
describe(name: string, database?: string): Description
5. search_kb - Search Knowledge Base
Search a knowledge base using hybrid semantic + keyword search.
search_kb(
kb_name: string,
question: string,
limit?: number, // default: 10
alpha?: number, // default: 0.5 (0=keyword, 1=semantic)
min_relevance?: number, // default: 0
metadata_filter?: Record<string, string>
): SearchResult[]
6. insert_kb - Insert Data into Knowledge Base
Insert data into a knowledge base from a source query.
insert_kb(
kb_name: string,
source_query: string, // e.g., "SELECT id, content FROM mydb.docs"
batch_size?: number, // default: 1000
threads?: number, // default: 1
skip_existing?: boolean // default: false
): { rows_inserted: number }
7. create_kb - Create Knowledge Base
Create a new knowledge base with embedding model configuration.
create_kb(
name: string,
embedding_model: {
provider: string, // openai, azure, google, ollama
model_name: string,
api_key?: string
},
content_columns: string[],
metadata_columns?: string[],
id_column?: string,
reranking_model?: {...},
storage?: string // default: chromadb
): { kb_name: string }
8. list_jobs - List Scheduled Jobs
List all scheduled jobs in MindsDB.
list_jobs(): Job[]
9. create_job - Schedule Recurring Query
Create a scheduled job to run a query at regular intervals.
create_job(
name: string,
query: string, // SQL to execute
every: string, // e.g., "1 hour", "1 day"
start?: string, // ISO datetime
end?: string // ISO datetime
): { job_name: string }
10. list_tables - List Tables
List all tables in a specific database.
list_tables(database: string): string[]
11. connect_database - Connect External Data Source
Connect an external database as a data source in MindsDB.
connect_database(
name: string,
engine: string, // postgres, mysql, mongodb, etc.
parameters: {
host: string,
port: number,
database: string,
user: string,
password: string
}
): { database_name: string }
Testing
With MCP Inspector
npx @modelcontextprotocol/inspector node dist/index.js
Or with a remote MindsDB instance:
MINDSDB_HOST=your-mindsdb-host MINDSDB_PORT=47335 \
npx @modelcontextprotocol/inspector node dist/index.js
Manual verification
After configuring Claude Code, restart and verify:
# In Claude Code, these tools should be available:
mcp__mindsdb__list_databases
mcp__mindsdb__list_knowledge_bases
mcp__mindsdb__search_kb
mcp__mindsdb__query
# ... etc.
MindsDB SQL Reference
Knowledge Base Query
-- Semantic search
SELECT * FROM my_kb WHERE content = 'search phrase' LIMIT 10;
-- With relevance filter
SELECT * FROM my_kb WHERE content = 'query' AND relevance >= 0.5;
-- Hybrid search (keyword + semantic)
SELECT * FROM my_kb WHERE content = 'query' AND hybrid_search_alpha = 0.8;
-- Metadata filter
SELECT * FROM my_kb WHERE content = 'query' AND metadata_column = 'value';
Output Columns
chunk_id:<id>:<chunk_number>of<total>:<start_char>to<end_char>chunk_content: The text contentrelevance: 0-1 score (higher = better match)distance: Vector distance (lower = closer)
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
E2B
Using MCP to run code via e2b.