Safely let AI query MS SQL Server in read-only mode.
This tool is described as an open-source, read-only MCP server for MS SQL Server, with no extra API keys and no declared remote egress endpoints; no clear high-risk red flags are evident. However, its core capability is querying a database, and its third-party provenance plus weak adoption/maintenance signals warrant cautious use with least privilege and isolation.
The material states there are no extra keys/environment variables, but as an MS SQL Server tool it would typically still rely on database-side credentials or trusted authentication. Misconfigured database access could expose query capability, which is a normal credential risk; no explicit abuse pattern is described.
No third-party remote endpoint is declared, and the material does not indicate data is sent to unrelated services. However, by design it must connect to an MS SQL Server and transmit queries/results to that database, which is ordinary network access consistent with its stated purpose.
The system flags that it executes code; as an MCP server, it would normally run a local Python process to provide tool functionality. The description claims 'no shell access, no stored procedures, read-only,' which would reduce the execution surface if true, but these constraints cannot be fully verified from the brief material alone.
Its core capability is reading data from MS SQL Server, so its access scope includes whatever database objects the configured account can read. Although it is described as read-only with no write capability, the practical risk depends on the actual privileges of the database account; no excessive system-level permissions beyond the stated function are indicated.
There is a public open-source GitHub repository, making the code in principle auditable, which is a positive signal. However, it comes from a third-party registry, has no declared license, shows 0 stars, and has unknown maintenance status, so supply-chain confidence is only moderate and code/dependency review is advisable.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "mssql-mcp-server-python" yet — see the docs or source repo.
Connect to my MS SQL Server in read-only mode and query the orders table. Return daily order counts and revenue for the last 30 days, sorted by date ascending.
A 30-day sales summary suitable for trend analysis or reporting.
Inspect the customers, orders, and order_items tables and list their columns, data types, and primary/foreign key relationships clearly.
A clear schema and relationship summary to help understand the database design.
In read-only mode, query the inventory table and find records with negative stock, null values, or no updates in the last 90 days, then provide grouped counts.
A list of anomalous records and summary counts for data quality checks.
Query and manage Microsoft SQL Server databases with natural language.
Securely connect AI assistants to SQL Server for queries, analysis, and admin tasks.
Run read-only SQL Server queries and inspect query performance statistics.
Connect AI to Microsoft SQL Server for schema queries and CRUD operations.
Let AI query multiple SQL databases read-only and export results.
Safely query multiple SQL databases in read-only mode across servers.