inaccurate evidence codes?

[kcs3]

In https://ctd2-dashboard.nci.nih.gov/dashboard/#observation/20190212-columbia-paracne-1-colony-0, I note a few possible problems the automatically assigned evidence codes (found with the question mark icons in the evidence type column:

In the 5th and 6th rows, which link out to submitter-assigned ECO codes, the automatically assigned code for evidence type "Link" is http://www.evidenceontology.org/term/ECO:0000212/, "combinatorial evidence". The term description is "A type of evidence in which at least two distinct types of evidence have been integrated.". This does not seem right for a generic link.

However, in looking through the ECO ontology terms, I do not find anything appropriate...

[kcs3]

suggest wild cards