Which stage in the aggregation pipeline is used to reshape the documents?

The $project stage in the aggregation pipeline is specifically designed to reshape documents by allowing you to include, exclude, or rename fields in the output. This stage enables you to transform the structure of the documents as they flow through the pipeline.

When using $project, you can create new fields by applying expressions or calculated values, adjust the format of existing fields, and eliminate fields that are not needed in the final output. For instance, if you have a document with several fields but only need a subset for a particular analysis, $project gives you the flexibility to specify exactly which fields to keep and how they should appear in the transformed document.

The other stages serve different purposes: $match filters documents based on criteria, $group aggregates data into groups based on specified keys, and $sort orders documents according to specified fields. Each of these stages is important in the aggregation process but does not directly focus on reshaping the document structure as $project does.