-The analyses performed for the sensitivity to the couplings are very intesting but also a very recent development. Did we check with our TH friends that we are using/ interpreting these tools/calculations correctly? Just to be on the safe side...

-Ref [44] is just a conference proceedings from May 2017. There is no published paper on that till now? It is always risky to have a our paper depending on a methodology that is unpublished. Usually there is a reason for the fact it did not appear on the arXiv yet as a paper to be peer reviewed .. It is of course physics management choice if they wish to take this risk in CMS.

- Do I understand we do not need to perform a re-analysis in this paper (just a rebinning where appropriate and a few other things like fixing the Z branching fractions, performing background model re-evaluation in the new bins..) Is that all what has changed w.r.t. the individual publications. If not it should be listed more clearly

- We use the H-> gamgam data from ref 30, but what is ref 30 exactly? Is this not a official CMS paper? Bizarre! I hope we did take it from a published paper, as required.

Details

- line 19: the result obtained in ref 13, was that with 8 TeV or 13 TeV data? One should add that in the text.

- line 20: I assume these are "one-sigma" bounds? Best to specify in the text.

- line 27: is that still the case when comparing to other methods mentioned above, which will of course also get more precise with increased luminosity?

- line 59: We mention nothing on the trigger in this section.. Normally we add a few lines on the L1/HLT in our papers.

- line 128-130: An influx of 1% from outside the fiducial volume to inside is very low, it seems. Probably I am missing something on what this number here exactly means...

- line 168: what "interference" is here meant exactly? Not between the final states.

- line 254: is this improvement consistent with the relatively increased of the data sample? I guess we have to compare the -4.3 < K < 4.3 result with the one of ref [13], right? Then the improvement looks much more than due to the increased data sample. What could be the reason for that? I guess you have investigated that.

- line 272 "in in"

- The references are sloppy! Totally incomplete references : 9,10,18,28,43 !!