dataoil.tech
MCP server to query Data Oil databases with SQL, schema discovery and search
dataoil.tech is an MCP server for databases. It runs locally on your machine and is started by your AI client.
- Platforms
- API
- Runs
- Locally
- Version
- 0.1.9
- Listing updated
Similar MCP servers
Install
Run in a terminal
claude mcp add mcp -- npx -y @dataoil/mcpAdd to claude_desktop_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@dataoil/mcp"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@dataoil/mcp"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@dataoil/mcp"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@dataoil/mcp"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@dataoil/mcp"
]
}
}
}Using several servers? Build one config file with the MCP config generator.
About dataoil.tech
Connects AI clients over stdio to Data Oil databases, so a model can read schemas, run SQL, search records and ingest data without leaving its chat or coding session. The registry entry lists SQL, schema discovery, search and ingestion as its three areas.
The README describes a public read-only sandbox that the sample projects sign into by default. Writes there are refused with a 403, while full keyword search is available. Using your own tenant needs an API key or an account name and password, and the README says creating databases requires a tenant administrator or Database Managers role.
The package is aimed at developers and data teams who already use Data Oil, or are evaluating it, and want an AI assistant to work against their databases. The README says the public clients are new, the source is closed, and semantic vector search is disabled in the free sandbox.
What it does
- SQL querying
- Runs SQL queries against Data Oil collections from an MCP-compatible AI client.
- Schema discovery
- Lists databases and reads collection schemas so the assistant can plan its queries.
- Keyword search
- Searches stored text by keyword, including across the sandbox's five-language passages.
- Data ingestion
- Writes records into a tenant database when the credential allows it.
- Read-only public sandbox
- Lets you try the server against a public sample database with a read-only account.
- Stdio transport
- Runs as a local process over stdio, installed from npm as @dataoil/mcp.
Good for
- Asking an AI assistant questions about a Data Oil database in plain language
- Exploring an unfamiliar collection's schema before writing queries
- Running keyword searches across stored documents from a coding assistant
- Testing an MCP integration against the read-only public sandbox before using a tenant
dataoil.tech FAQ
How is the server installed?
The registry entry names the npm package @dataoil/mcp at version 0.1.9, which runs over stdio. The README gives the command npm install @dataoil/mcp@0.1.9.
Can the public sandbox be written to?
No. The README says every write to the public sandbox is refused with 403 Security.
What credentials does a tenant need?
The README says an API key, minted in the console or with POST /api/v2/tokens, or an account name and password is needed for your own tenant.
Is the source code open?
The README says the public clients are new and their source is closed. The package is covered by the Data Oil Software Licence.
How do I add dataoil.tech to Claude Code?
Run this command in your terminal: claude mcp add mcp -- npx -y @dataoil/mcp
What are the best alternatives to dataoil.tech?
The closest alternatives to dataoil.tech are SQL Server MCP (Windows), MCP Database Server, stoolap, Mongodb MCP Server and SynapCores.
Where this comes from
- Description: written from the project's README and its MCP Registry entry.
- Install steps: from the official MCP Registry entry.