mssql-mcp-server
A Microsoft SQL Server MCP server that enables querying, inspecting, and troubleshooting SQL Server instances via ODBC with per-caller credential elicitation.
README
mssql-mcp-server
A Microsoft SQL Server MCP server. Query, inspect, and troubleshoot any SQL Server instance from Cursor, Claude Code, Codex, or any MCP (Model Context Protocol) client. Connects via ODBC using per-caller credentials elicited at runtime. Passwords live only in server memory with an idle TTL and are never logged.
- Query tools:
connect/disconnect/list_sessionssession lifecycle,execute_sqlwith table/JSON/CSV output, andexplain_queryfor estimated or actual Showplan XML execution plans - Per-caller identity: the
X-DB-UserandX-DB-Passwordheaders are both optional; anything missing is elicited once and cached in-memory - Permission-scoped access: read vs. write is governed by the SQL login's own permissions;
provision_mcp_read.sqlprovisions a ready-made read-onlymcp_readlogin across all user databases - Zero secrets on disk: no credentials in config files, env vars, or logs
Example prompts
- "Show the execution plan for this query and suggest a better index."
- "Who is blocking whom on
db1right now?" - "List the 10 largest tables in the
salesdatabase." - "Find all stored procedures that reference the
Orderstable." - "Export the results of this query as CSV."
Tools
| Tool | Description |
|---|---|
connect |
Open a session to a SQL Server database and return a session_id |
disconnect |
Close an active SQL Server session |
list_sessions |
List active sessions with server, database, and usage details |
execute_sql |
Run a SQL statement and return rows as table, JSON, or CSV |
explain_query |
Estimated or actual execution plan as Showplan XML |
Quick Start
docker compose -f docker-compose.yml -p mssql-mcp up -d --build --force-recreate
Client setup
Both X-DB-User and X-DB-Password headers are optional: anything missing is prompted for once via MCP elicitation and cached in memory.
Cursor (mcp.json)
{
"mcpServers": {
"mssql-mcp": {
"type": "http",
"url": "http://localhost:8007/mcp",
"headers": {
"X-DB-User": "<your-db-username>",
"X-DB-Password": "<your-db-password>"
}
}
}
}
Claude Code
claude mcp add --transport http mssql-mcp http://localhost:8007/mcp \
--header "X-DB-User: <your-db-username>" \
--header "X-DB-Password: <your-db-password>"
Codex (~/.codex/config.toml)
[mcp_servers.mssql-mcp]
url = "http://localhost:8007/mcp"
http_headers = { "X-DB-User" = "<your-db-username>", "X-DB-Password" = "<your-db-password>" }
Any other MCP client works the same way: point it at the streamable HTTP endpoint and pass the headers.
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.
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.
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.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.