SAP HANA MCP Server
MCP server for SAP HANA and HANA Cloud: schema browsing, SQL with row caps and 34 tools, with writes blocked by default
- 75 ★ on the repository
- Last commit
Broad access
Why this level
- The server connects to a live SAP database with the login and password from the agent config
- Enabling `HANA_ALLOW_UPDATE` or `HANA_ALLOW_DELETE` gives the agent the right to change and delete data
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add mcp/sap-hana-mcp-serverDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- The server connects to a live SAP database with the login and password from the agent config
- Enabling `HANA_ALLOW_UPDATE` or `HANA_ALLOW_DELETE` gives the agent the right to change and delete data
This is third-party code. Review the repository files before installing.
What it does
SAP HANA MCP Server starts with `npx -y hana-mcp-server` and connects an agent to a SAP HANA or HANA Cloud database over the SQL port. It has 34 tools. Some list schemas, tables, columns, indexes, foreign keys, views and DDL. Others run parameterized SQL with caps on rows, columns and cells and page through large result sets. Further tools collect table and column statistics and show execution plans, dependencies, partitions and the most expensive queries. INSERT, UPDATE, DELETE and TRUNCATE are blocked, and each one is enabled by its own environment variable: `HANA_ALLOW_INSERT`, `HANA_ALLOW_UPDATE`, `HANA_ALLOW_DELETE`. If you add a JSON file describing the meaning of tables and code values, `hana_explain_table` returns that description to the agent alongside the structure. The default transport is stdio; the HTTP variant with Bearer JWT checks starts from a clone of the repository with `npm run start:http`.
Who it is for. Developers and analysts working on SAP HANA who want to read a database schema and data from Claude Code, Cursor or Cline without hand-written queries.
Good fit when
- An agent has to understand an unfamiliar HANA schema: its tables, columns, relations and indexes
- A HANA query is slow and you need the execution plan and a list of the most expensive queries
- You want sample data and column statistics before writing a report
- An SAP ABAP database is full of code values and the agent should read them through a table-semantics description
Not a fit when
- You need a different database: the server works only with SAP HANA and HANA Cloud
- The agent has to change data in bulk: writes are blocked by default and enabling them on a production database without a copy is unwise
- You need an interface for people without access to the SQL port: the server connects straight to the database port
Example request
List all tables in schema SAPABAP1 with ORDER in the name and describe the columns of the orders tableMCP
- Transport
- stdio
- Authentication
- API key
Environment variables
HANA_HOSTrequired- HANA server host
HANA_PORTrequired- Database SQL port
HANA_USERrequired- Database user
HANA_PASSWORDrequired, secret- User password
HANA_SCHEMA- Default schema
Limitations
Node.js 18 and access from your machine to the HANA SQL port are required. The password and other connection settings live in the client environment variables, so they end up in the agent config. Version 0.3.1 was verified by the author on HANA Cloud; local multi-tenant installs set `HANA_DATABASE_NAME`. The npx package works over stdio only: the HTTP mode is available from a clone of the repository. Table-semantics descriptions are meant to be written by hand as JSON and are not generated automatically. Changing environment variables needs a restart of the server connection in the client.
How to disable. Remove the `hana` entry from `.mcp.json` or `~/.claude.json`, or run `claude mcp remove hana`, and restart the client. If you installed it globally, run `npm uninstall -g hana-mcp-server`.
FAQ
How do I connect HANA MCP Server to Claude Code?
Add an entry to `.mcp.json` with type stdio, command `npx`, args `-y hana-mcp-server` and an env block with the connection settings. Restart the server connection afterwards. The block is the same for Cursor, Cline, Windsurf and VS Code.
Can the agent delete data in HANA?
Not by default: INSERT, UPDATE, DELETE and TRUNCATE are blocked. Each operation is allowed by its own variable, for example `HANA_ALLOW_DELETE`. Until you enable them the server only reads.
What does the hana_explain_table tool do?
It returns the table structure and, if you supplied a semantics JSON, descriptions of the table, columns and code-value meanings. That lets the agent read SAP codes as words.
Is there an HTTP mode?
Yes, from a clone of the repository: `npm run start:http` listens on `127.0.0.1:3100` at `/mcp`. For anything beyond localhost, enable Bearer JWT checks and put the service behind a reverse proxy.
What is hana-mcp-ui?
A separate package: `npx hana-mcp-ui` opens a web interface for editing environment variables and deploying the setup to Claude Desktop.
Discussion
No questions or reviews yet.
Related
MCP serversSAP HANA CLI: MCP for working with SAP HANA
SAP HANA CLI
A SAP HANA developer CLI with an MCP server: an agent works with the database and HDI containers through core tools
One MCP server for SQLite, SQL Server, PostgreSQL and MySQL: read and change data, edit schemas and export to CSV and JSON
MCP server that gives an agent access to MySQL: schema inspection and queries, read-only by default, writes behind flags
MCP server for SAP BW/4HANA and BW 7.5 on HANA: an agent reads and changes modeling objects, runs queries and watches data loads.
A Python MCP server for MySQL: listing tables, reading data, running SQL and inspecting schema through a controlled interface
MCP MongoDB Server (kiliczsh)
MCP MongoDB Server
An MCP server that gives an agent queries, aggregations and writes in MongoDB and starts with one npx command, no install
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| SAP HANA MCP Server (this card) | no | API key | 75 | |
| SAP HANA CLI: MCP for working with SAP HANA | yes | not required | 115 | a SAP HANA developer tool with a CLI and MCP, more about HDI containers |
| MCP SAP Docs | no | not required | 252 | searches SAP documentation and does not connect to a database |
README in short
The README describes a server for SAP HANA and HANA Cloud: schema, guarded SQL and an optional business-meaning layer. Documentation is split into an environment reference, connection profile samples and notes on local HTTP. The capability list names 34 tools in six areas.