To positively identify images during import we will need an image digest. Additionally, downstream of the import process will require derivative information from the manifest. In order to achieve this stereoscope will need to surface more information:

• A manifest (raw or computed) : since stereoscope is using github.com/google/go-containerregistry it will not be possible in all cases to obtain a raw manifest --that is, the library is centered around the OCI v1 schema, which means requesting the v1image.RawManifest() really means computing it from the underlying docker-specific manifest.json. This is different with OCI images since the raw manifest can simply be read from the source. This implies that stereoscope should not include this information on the default execution path, but make it optionally available for import requests only (as computing the manifest can be costly).
• A digest (raw or computed) : Once we have the manifest the digest should be relatively simple to compute (a sha256 of the manifest, made available from v1image.Digest() (again, if the manifest isn't available, this is a costly process).
• An ID: we should fix an issue in stereoscope where the image.Manifest.Digest is really the ImageID not the digest (we should rename this field).