- SCHEMA
- INDEXING
- VALIDATION
- HUMAN
Short answer
How do I create vector tables and indices using my AI?
You describe the table structure and index requirements in plain English. Your AI then calls the MyScale API to build the tables and deploy the indices for you. You get a ready-to-query vector database without writing SQL.
The outcomes
Where your configuration lands
Your AI handles the heavy lifting of database configuration.
SCHEMA
Automated table creation
The AI translates your descriptions into a structured vector table. It handles the column definitions and dimension settings.
INDEXING
Specialized index deployment
Your agent sets up the specific indices needed for fast similarity searches. This ensures your vector data is searchable immediately.
VALIDATION
Configuration check
The AI confirms the table and index are live. You receive confirmation that the database is ready for data ingestion.
HUMAN
Manual schema review
You decide if the generated dimensions and index types match your specific model requirements. The AI waits for your approval before finalizing complex setups.
The workflow
What your AI does when the request arrives.
The AI acts as a bridge between your intent and the MyScale API.
Parse requirements
Your agent reads your natural language request to identify the necessary dimensions and vector types.
create_vector_tableBuild the table
The AI executes the command to initialize the vector storage with your specified parameters.
create_vector_tableApply indices
The agent deploys the specialized indices required for high-performance vector search.
create_vector_indexVerify deployment
The AI checks the MyScale environment to ensure the table and index are active and ready.
create_vector_index
Try it
Copy these to start.
Use these prompts to kick off your database setup.
Starting points
Swap out the dimension counts and index types to match your specific embedding model.
MyScale Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_vWeauVyztoYzGBsy6h9azMYGgJ3R8tQDDZfTvuoy/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — MyScale capabilities are ready to use.
{
"mcpServers": {
"myscale-sql-vector-database-api-mcp": {
"url": "https://edge.vinkius.com/vk_preview_vWeauVyztoYzGBsy6h9azMYGgJ3R8tQDDZfTvuoy/mcp"
}
}
}Create a new vector table for my text embeddings with 1536 dimensions.
Set up a specialized vector index for the 'user_embeddings' table.
Build a vector table named 'product_vectors' and add a search index to it.
Initialize a new table for image embeddings and deploy an index for fast retrieval.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect MyScale once, then ask.
One click connects your AI client to MyScale. Your credentials stay encrypted and secure on our side.
Connect MyScale to your AIFAQ
How this task behaves
- 01
Can the AI delete my existing tables?
No. The current tools only allow for creating new tables and indices.
- 02
Do I need to write SQL to create these tables?
No. You use natural language, and the AI handles the API calls to MyScale.
- 03
Can the AI manage my data rows?
No. This specific setup task is limited to creating the table structure and the indices.
- 04
Will the AI choose the index type for me?
The AI will suggest an index based on your description, but you should verify it fits your search needs.
- 05
Can I change the table dimensions after the table is created?
No. Once the table is created via the tool, you would need to handle structural changes manually.
More questions about MyScale? See everything the MyScale Connector can do
Connect MyScale to Claude, Cursor, ChatGPT & more
