Development Tools

HPXML Toolbox

To facilitate the adoption of the HPXML standard use cases, NREL has developed the HPXML Toolbox. Features of the toolbox include a data dictionary/schema explorer tool as well as a validator.

Data Dictionary and Schema Explorer

One of the challenges of HPXML is understanding and communicating the structure of the schema, how data elements are organized, and how they are to be interpreted. Up until recently the best way to learn those details was to inspect the schema files. The online data dictionary is a searchable website that documents HPXML structure, element interpretations, the standard datasets to which an element belongs, and how any element in HPXML relates to the other data standards such as BEDES and RESO.


The toolbox provides a schema validator plus standard dataset validator. Submitting an HPXML file will perform the following steps:

  1. Validation of the file against HPXML schema and determination of schema version.
  2. Validation against each of the standard use cases including output describing any missing elements required to meet a particular use case.
  3. Display of HPXML file in collapsable tree form.

The validation functionality is available as a website for testing and viewing individual files, or as a web service API that allows the validation to be called as part of a software workflow.

Data Selection Spreadsheet

A HPXML Data Selection Tool has been developed to help program administrators select the data points they are requiring to collect for specific use cases. The data selection tool is a spreadsheet with the required and optional data points selected by working group members for specific standard datasets, including Audit, Upgrade, Home Energy Score, and Home Performance Certificate. A detailed desccription of these use cases and the corresponding standard HPXML datasets can be found in section Standard HPXML Datasets.

In the spreadsheeet, data points required by a dataset will be in black color instead of light grey. Elements that are optional are highlighted in red.

The standard datasets provide a complete list of data elements required to either provide or process.