Doris-MCP-Lite

- Executeread-only SQL queriesagainst your Doris database.
- Performdata analysis operationssuch as retrieving yearly, monthly, and daily usage data.
- Query metadata such asdatabase schemas,table structures, andresource usage.
- Connection Pooling: Efficientconnection management with poolingto optimize performance.
- Asynchronous Execution: Support forasynchronous queryexecution to improve responsiveness.

- Built-in prompt templates to assist LLMs in askinganalytics questions.
- Support formulti-role promptingto enhance the interaction between LLMs and the Doris database.
- Support foruser-definedandgeneral-purposeSQL analysis prompts.

- Expose your Dorisdatabase schemaasstructured resources.
- Allow LLMs tocontextually accesstable and field definitions to improve query understanding.

Recommended for personal and server deployment

The script will automatically install the server and help you walk through database configuration.

✅ After installation, the command-line tool server will be available to launch the MCP server.

Option 3: Clone the source and install manually

Recommended if you want to modify the server

git clone https://github.com/YOUR_USERNAME/doris-mcp-lite.git cd doris-mcp-lite

uv venv # Create a virtual environment uv sync # Install dependencies # Activate the virtual environment source .venv/bin/activate # On Windows: .venv\Scripts\activate uv pip install

- Add this server to your LLM client or Run the server:

uv run server doris://user:pass@localhost:9030/mydb

uv pip install 'git+https://github.com/NomotoK/doris-mcp-lite.git' uv sync source .venv/bin/activate # On Windows: .venv\Scripts\activate uv pip install -e . uv run server doris://user:pass@localhost:9030/mydb

Use the .env file to permanently save your database connection information in the MCP server, so you do not need to enter the database connection every time you run the MCP server with CLI. Of course,this step is not necessary, if you are using a MCP-capatible LLM client, you can alsoset up a database connection in the configuration fileof the MCP client later (See step2). Please follow these steps to finish configuration:

After installing, navigate to thedoris_mcp_lite/config/directory inside your project directory. If you are using pip, your package will be installed in Python site-packages:

-

Mac/Linux:/Users/YOUR_USERNAME/.local/lib/python3.x/site-packages/doris_mcp_lite/config/

Windows:C:\Users\YOUR_USERNAME\AppData\Local\Programs\Python\Python3x\Lib\site-packages\doris_mcp_lite\config\

You can run the following command to locate pip install location:
- Edit .env to set yourDorisdatabase connection information:

DB_HOST=your-doris-host DB_PORT=9030 DB_USER=your-username DB_PASSWORD=your-password DB_NAME=your-database MCP_SERVER_NAME=DorisAnalytics DEBUG=false

✅ Now your LLM client will discover Doris Analytics tools, prompts, and resources through the MCP server.

Before you start, you can run thetest.pyin the projectsrc/doris-mcp-litedirectory to directly call the MCP Server functional interface to test database connection, resources, tools, etc. without using LLM (such as Claude, GPT, etc. models). You can control what functions to test by passing arguments through the command line.

Test all resources exposed by the server:

python test.py --server server.py --test resources

or test all the tools provided by the server:

python test.py --server server.py --test tools

python test.py --server "doris://user:pass@localhost:9030/mydb" --test dbconfig

or test all functions of resources, tools, and prompt words at one time:

python test.py --server server.py --test all

Testing Database connection and run server

Launch the MCP server by running the command:

server doris://user:pass@localhost:9030/mydb

python -m doris_mcp_lite.server doris://user:pass@localhost:9030/mydb

The server immediately attempts to connect to the database. If the connection is successful, after startup, you should see:

Contributions are welcome! Feel free to open issues or submit pull requests.

Hydrolix time-series datalake integration providing schema exploration and query capabilities to LLM-based workflows.

614 million contacts, live SMTP-verified. Invalid emails cost you $0. Connect Argorant with any Agent through CLI, MCP or API and build verified lists in Claude, GPT or Hermes.

Chinese writing toolbox for Claude Code: text analysis, outline generation, GB/T 7714 references,character frequency

C# SQL Agent MCP server featuring raw SQL input, strict AST validation, and an embedded Admin UI. Eliminates LLM hallucinations and security risks across 6 major databases.

Build and access a personal LLM knowledge base from your editor or client without any infrastructure setup.

Mantora is a local-first MCP observer: a lightweight UI + proxy for inspecting LLM data access (sessions, tool calls, results) with protective defaults.

Biologically-inspired memory for Claude. Hebbian plasticity, dream cycles, and temporal decay for persistent knowledge synthesis.

Access the Quran API from alquran.cloud to retrieve accurate Quranic text and reduce LLM hallucinations.

An MCP server for local Tabular Models like PowerBI. It allows LLM clients to debug, analyze, and compose DAX queries by connecting to a local Tabular model instance.