This page gives an intro on how to get started with the Toolchest client.

First, make sure that:

  • The Toolchest client is installed, which you can do via:

    $ pip install toolchest-client
  • The Toolchest client is up-to-date (be sure to use the latest version, currently 0.7.13)

  • You have an authentication key

Configuring Your Client

Toolchest requires an authentication key in order to process queries.

You can obtain a key by signing up for a trial of Toolchest at this link.

Contact Toolchest if:

  • you’ve forgotten your key

  • the key is producing authentication errors.

Once you have an authentication key, configure your client as follows:

>>> import toolchest_client as toolchest
>>> toolchest.set_key(YOUR_TOOLCHEST_KEY)

The variable YOUR_TOOLCHEST_KEY can be either a string containing your key or a path to a file containing the key (and nothing else).

Sample Data

To use Toolchest tools, you will need input files.

A sample input file (~50MB) can be downloaded at this link.

Using Tools

To use a data analysis tool, you would probably run a command like this on the command line:


Once Toolchest is configured, you can do the same with the corresponding Toolchest function:

>>> import toolchest_client as toolchest
>>> toolchest.your_tool(
...     tool_args="YOUR_CUSTOM_TOOL_ARGS",
...     inputs="path/to/input",
...     output_path="path/to/output",
... )

Here, your_tool is the name of the command that you would use, and YOUR_CUSTOM_TOOL_ARGS is a string containing all the arguments that you would normally pass to the tool, outside of input and output filepath arguments.

Note that YOUR_CUSTOM_TOOL_ARGS should not include any arguments related to the input and output file paths; these will be automatically handled by the Toolchest backend, and including these arguments will lead to undesired output.

For a full list of available tools, see this list.

Not all functionalities of available tools can be used. See the relevant in-depth tool documentation or contact Toolchest for details.