Python is initially only aware of tools stored in ArcGIS system toolboxes like the Data Management Tools, Conversion Tools, and Analysis Tools toolboxes. Custom tools created by an individual, third party, or organization and stored in a custom toolbox can be accessed in the Python window like any system tool by importing the custom toolbox into the ArcPy site package.
In the following example, the ImportToolbox function is used to allow tools contained in a custom toolbox to be accessed in Python. After importing the toolbox, the custom tools can be accessed as arcpy.<toolname>_<alias>.
>>> arcpy.ImportToolbox("c:/mytools/geometrytools.tbx")
>>> arcpy.CreateRegularPolygons_geometry(When a tool is accessed through the ArcPy site package, the toolbox alias where the tool is contained is a required suffix (arcpy.<toolname>_<alias>). Since ArcPy depends on toolbox aliases to access and execute the correct tool, aliases are extremely important when importing custom toolboxes. A good practice is to always define a custom toolbox's alias. However, if the toolbox alias is not defined, a temporary alias can be set as the second parameter of the ImportToolbox function.
>>> arcpy.ImportToolbox("c:/mytools/geometrytools.tbx", "mytools")
>>> arcpy.CreateRegularPolygons_mytools(Server toolboxes
Geoprocessing services can also be added to the scripting environment using ImportToolbox. Whether adding the geoprocessing service from a local or Internet server, the server and toolbox name are semicolon delimited.
Sample syntax for adding a geoprocessing service
# Import a geoprocessing service
import arcpy
# To add a toolbox from a Internet server, provide the url and toolbox name
# delimited by a semi-colon
arcpy.ImportToolbox("http://lab13/arcgis/services;BufferByVal")Sample syntax for adding a local geoprocessing service
# Import a local geoprocessing service
import arcpy
# To add a toolbox from a local server, provide the server and toolbox name
# delimited by a semi-colon
arcpy.ImportToolbox("lab13;BufferByVal")