Alibaba Cloud DMS MCP Server

Instructions

Implementation Reference

• src/alibabacloud_dms_mcp_server/server.py:455-479 (handler)

  The handler function `nl2sql` that implements the core logic of the `generateSql` tool by calling the Alibaba Cloud DMS Enterprise API `generate_sql_from_nl` to convert natural language to SQL.
  async def nl2sql( database_id: str = Field(description="DMS databaseId"), question: str = Field(description="Natural language question"), knowledge: Optional[str] = Field(default=None, description="Optional: additional context"), model: Optional[str] = Field(default=None, description="Optional: if a specific model is desired, it can be specified here") ) -> SqlResult: client = create_client() req = dms_enterprise_20181101_models.GenerateSqlFromNLRequest(db_id=database_id, question=question) if knowledge: req.knowledge = knowledge if mcp.state.real_login_uid: req.real_login_user_uid = mcp.state.real_login_uid if model: req.model = model try: resp = client.generate_sql_from_nl(req) if not resp or not resp.body: return SqlResult(sql=None) data = resp.body.to_map() sql_content = data.get('Data', {}).get('Sql') if data else None return SqlResult(sql=sql_content) except Exception as e: logger.error(f"Error in nl2sql_explicit_db: {e}") raise
• src/alibabacloud_dms_mcp_server/server.py:748-749 (registration)

  Registration of the `generateSql` tool, binding the `nl2sql` handler function to the MCP tool named `generateSql`.
  self.mcp.tool(name="generateSql", description="Generate SELECT-type SQL queries from natural language input.", annotations={"title": "自然语言转SQL (DMS)", "readOnlyHint": True})(nl2sql)
• src/alibabacloud_dms_mcp_server/server.py:117-119 (schema)

  Pydantic model defining the output schema for the `generateSql` tool, containing the generated SQL string.
  class SqlResult(MyBaseModel): sql: Optional[str] = Field(description="The generated SQL query")