What it does

Connects an agent to Microsoft SQL Server: inspect tables, run queries, read results. The schema view alone saves an afternoon on an unfamiliar database. Use a read-only login — the connection string is the entire permission model, and it is easy to hand over a login with more rights than the task needs.

Install

Terminal

npx -y mssql-mcp-server

Client configuration

Paste this into your MCP client config (Claude Desktop, Cursor, or any client that reads mcpServers), then restart the client.

mcp.json

{
  "mcpServers": {
    "mssql": {
      "command": "npx",
      "args": [
        "-y",
        "mssql-mcp-server"
      ]
    }
  }
}

Use it for

Exploring schema and running read-only queries against SQL Server

Skip it if

Production credentials with write scope

When to use it

Reach for this when the task needs database inside the agent loop. Skip it if a plain shell command or a one-off script is enough — every server you add costs context and adds a failure mode.

Other TypeScript servers

Filesystem — Gives an agent scoped read, write, and search access to directories you explicitly allow
Memory — A persistent knowledge graph stored in a local JSON file, exposed through entities, relations, and observations
Sequential Thinking — Gives the model a scratchpad for structured, revisable reasoning: numbered thoughts that can branch, revise, and conclude

MSSQL MCP