It sounds like you want something similar to what "Use measured retention times when present" does, except that you want Skyline to pay attention to the start and end times of the peak boundaries instead of using a fixed number of minutes from the apex of that peak.
Unfortunately, I don't think there's any straightforward way to change how Skyline determines those window widths.

You might be able to achieve what you want by pasting the correct values into the "Explicit Retention Time" and "Explicit Retention Time Window" columns for all of your molecules.

That is, you would create a report that includes the "Min Start Time" and "Max End Time" columns, and then export that report to a .csv file. Then, in something like Microsoft Excel you would do some calculations to decide what the "Explicit Retention Time" and "Explicit Retention Time Window" values should be. "Explicit Retention Time" might be "(MinStartTime + MaxEndTime) / 2" and "Explicit Retention Time Window" might be "1 + (MaxEndTime - MinStartTime) / 2" in order to have a window which starts 1 minute before MinStartTime and ends 1 minute after MaxEndTime.
Then, you would copy all those calculated values from Excel and paste them into a report that has the "Explicit Retention Time" and "Explicit Retention Time Window" columns.

Skyline does not have any features for changing the predicted retention time window by clicking on the chromatogram. The only thing you can adjust by clicking on the chromatogram are the integration boundaries.

Note that after you have exported your method you should probably blank out all the Explicit Retention Time values before you import chromatograms because the having an Explicit Retention Time set often negatively impacts Skyline choosing the correct peak.
-- Nick

Thanks for the fast reply and you have understood my question correctly.
 I get a value of start and stop time for my integrated peak when I integrated it. Those are the values I want to use when I export my instrument method for data acquisition. This would allow different width for my compound instead if the fixed value around the peak apex and I could select these depending on the peak shapes.

As you say I can export the start and stop time for peak integration to csv and calculate the variables in excel then paste the new values back (as explicit RTs) to skyline this worked as you suggested.

It is a simple calculation that I would appreciate to have built into Skyline for the export method (it would be faster than the export and import to excel). Would be nice if this would be possible but understand you have to prioritize between all requests and the export to all venodor files might be lots of work. My request is for Waters TQ instrument files

Just wanted to add, being able to run a SST sample with all compounds at the start, integrating each peak to set MRM aquistion start and stop and export this as a scheduled method seems like simple thing to implement that would give the optimal scheduling with a very fast step. Please look into this we must be many that would benefit from this as we need to maximize dwell times!