FOCUS DATA MCP Server [中文]
A Model Context Protocol (MCP) server enables artificial intelligence assistants to convert natural language into SQL statements.
There are already so many Text-to-SQL frameworks. Why do we still need another one?
In simple terms, focus_mcp_sql adopts a two-step SQL generation solution, which enables control over the hallucinations of LLM and truly builds the trust of non-technical users in the generated SQL results.
Below is the comparison table between focus_mcp_sql and others:
Comparison Analysis Table
Here’s a side-by-side comparison of focus_mcp_sql with other LLM-based frameworks:
Feature | Traditional LLM Frameworks | focus_mcp_sql |
---|---|---|
Generation Process | Black box, direct SQL generation | Transparent, two-step (keywords + SQL) |
Hallucination Risk | High, depends on model quality | Low, controllable (keyword verification) |
Speed | Slow, relies on large model inference | Fast, deterministic keyword-to-SQL |
Cost | High, requires advanced models | Low, reduces reliance on large models |
Non-Technical User Friendliness | Low, hard to verify results | High, easy keyword checking |
Features
-Initialize the model -Convert natural language to SQL statements
Prerequisites
- jdk 23 or higher. Download jdk
- gradle 8.12 or higher. Download gradle
- register Datafocus to obtain bearer token:
- Register an account in Datafocus
- Create an application
- Enter the application
- Admin -> Interface authentication -> Bearer Token -> New Bearer Token
Installation
- Clone this repository:
- Build the server:
MCP Configuration
Add the server to your MCP settings file:
Available Tools
1. gptText2sqlStart
initial model.
Parameters:
model
(required): table modelbearer
(required): bearer tokenlanguage
(optional): language ['english','chinese']
Example:
model 参数说明:
名称 | 位置 | 类型 | 必选 | 说明 |
---|---|---|---|---|
model | body | object | 是 | none |
» type | body | string | 是 | 数据库类型 |
» version | body | string | 是 | 数据库版本 |
» tables | body | [object] | 是 | 表结构列表 |
»» tableDisplayName | body | string | 否 | 表显示名 |
»» tableName | body | string | 否 | 表原始名 |
»» columns | body | [object] | 否 | 表列列表 |
»»» columnDisplayName | body | string | 是 | 列显示名 |
»»» columnName | body | string | 是 | 列原始名 |
»»» dataType | body | string | 是 | 列数据类型 |
»»» aggregation | body | string | 是 | 列聚合方式 |
» relations | body | [object] | 是 | 表关联关系列表 |
»» conditions | body | [object] | 否 | 关联条件 |
»»» dstColName | body | string | 否 | dimension 表关联列原始名 |
»»» srcColName | body | string | 否 | fact 表关联列原始名 |
»» dimensionTable | body | string | 否 | dimension 表原始名 |
»» factTable | body | string | 否 | fact 表原始名 |
»» joinType | body | string | 否 | 关联类型 |
2. gptText2sqlChat
Convert natural language to SQL.
Parameters:
chatId
(required): chat idinput
(required): Natural languagebearer
(required): bearer token
Example:
Response Format
All tools return responses in the following format:
Visual Studio Code Cline Sample
- vsCode install cline plugin
- mcp server config
- use
- initial model
- transfer: what is the max age
- initial model
Contact:
This server cannot be installed
remote-capable server
The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.
A NL2SQL plugin based on FocusSearch keyword parsing, offering greater accuracy, higher speed, and more reliability!
Related Resources
Related MCP Servers
- AsecurityAlicenseAqualityEnables querying documents through a Langflow backend using natural language questions, providing an interface to interact with Langflow document Q\&A flows.Last updated -114JavaScriptMIT License
- AsecurityAlicenseAqualityUsed to create an assistant integrated with n8n that can search n8n documentation, example workflows, and community forums.Last updated -110PythonMIT License
- -securityFlicense-qualityConnects to Cursor and enables deep web searches via Linkup and RAG capabilities using LlamaIndex.Last updated -1Python
- AsecurityFlicenseAqualityA lightweight toolkit that enables Claude to search Twitter with natural language queries and display results based on user intent, supporting features like tweet filtering, pagination, and flexible output formatting.Last updated -18JavaScript