create_database
✓ SAFE: Create a new database in your Turso organization. Database name must be unique.
LABSCO SUMMARY
The destructive half of a database client is a different tool, not a different argument.
Reading is its own entry point: execute_read_only_query accepts SELECT, PRAGMA and EXPLAIN and rejects write operations outright, so an agent that should only read can be handed that tool and cannot get past it by phrasing the SQL differently. The genuinely destructive paths — execute_query and delete_database — are labelled as such in their own descriptions and ask the caller to confirm first, which is a convention in the text rather than a lock. The token you configure is organization-wide, so what the server can reach is decided when you set it, not per call.
What it is
A two-level Turso Cloud server: organization tools that create, list and delete databases and mint tokens for them, and database tools that read schema, run SQL and search vectors.
What you get
- Databases created, listed and deleted across your Turso organization, with group and regions set at creation
- Per-database tokens minted with a chosen permission by generate_database_token
- Schema on demand — list_tables for a database, describe_table for one table
- SQL split by risk: execute_read_only_query takes SELECT, PRAGMA and EXPLAIN and rejects write operations, while execute_query is the one that can INSERT, UPDATE, DELETE, DROP and ALTER
- Vector similarity search over a named vector_column, given a query_vector and a limit
Requirements
A Turso account with an API token in TURSO_API_TOKEN, plus TURSO_ORGANIZATION and TURSO_DEFAULT_DATABASE so calls that omit a database still have one to run against.
Setup effort
One command plus a key — npx -y mcp-turso-cloud, then supply credentials
USE CASES
01
Standing up a throwaway database for a feature branch
Create the database in the right group and regions, mint a token scoped to it, and confirm it appears in the organization before pointing anything at it.
02
Answering a question about production data without risking it
See which tables exist, read the columns of the one you need, then run the SELECT through the tool that refuses anything that would write.
03
Finding the rows nearest to an embedding you already have
Point the search at the table and the column holding the vectors, pass the query vector, and cap how many neighbours come back.
9 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
delete_database
⚠️ DESTRUCTIVE: Permanently deletes a database and ALL its data. Cannot be undone. Always confirm with user before proceeding and verify correct database name.
describe_table
Gets schema information for a table
execute_query
⚠️ DESTRUCTIVE: Execute SQL that can modify/delete data (INSERT, UPDATE, DELETE, DROP, ALTER). Always confirm with user before destructive operations.
execute_read_only_query
✓ SAFE: Execute read-only SQL queries (SELECT, PRAGMA, EXPLAIN). Automatically rejects write operations.
generate_database_token
Generate a new token for a specific database
list_databases
List all databases in your Turso organization
list_tables
Lists all tables in a database
vector_search
Performs vector similarity search
BEFORE YOU INSTALL
What to know first
Tools you get
9 tools when we connected. execute_read_only_query accepts SELECT and PRAGMA; execute_query runs INSERT, UPDATE, DELETE, CREATE and DROP, and is meant to require explicit approval in your client.
Requires
- npx on PATH
- TURSO_API_TOKEN in your environment
- TURSO_ORGANIZATION in your environment (your Turso organization name)
CONTEXT COST
Light Standing cost
Rated by the size of the 9 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Connect your client
Server setup
TERMINAL
npx -y mcp-turso-cloudMCP Clients
Cline, Claude Desktop
JSON
{
"mcpServers": {
"mcp-turso-cloud": {
"command": "npx",
"args": [
"-y",
"mcp-turso-cloud"
],
"env": {
"TURSO_API_TOKEN": "<YOUR_TURSO_API_TOKEN>",
"TURSO_ORGANIZATION": "<YOUR_TURSO_ORGANIZATION>",
"TURSO_DEFAULT_DATABASE": "optional-default-database"
}
}
}
}Claude Desktop — WSL
Inside WSL, clone the repository, install its dependencies and build it:
CLAUDE DESKTOP — WSL · TERMINAL
git clone https://github.com/spences10/mcp-turso-cloud
cd mcp-turso-cloud && npm install && npm run buildThen add this to your Claude Desktop configuration; <YOUR_PATH_TO_MCP_TURSO_CLOUD> is the path of that clone inside WSL:
CLAUDE DESKTOP — WSL · JSON
{
"mcpServers": {
"mcp-turso-cloud": {
"command": "wsl.exe",
"args": [
"bash",
"-c",
"TURSO_API_TOKEN=<YOUR_TURSO_API_TOKEN> TURSO_ORGANIZATION=<YOUR_TURSO_ORGANIZATION> node <YOUR_PATH_TO_MCP_TURSO_CLOUD>/dist/index.js"
]
}
}
}
