Wireframing, a declining trend?

As an interaction designer, one of the job requirements is always wireframing. Lately I've seen the trend towards skipping the wireframing stage and going straight to a clickable prototype. 

For me, this breaks down to several things. A) Is this a new app? B) Is this a redesign? C) Is it mobile?

New applications

I would have to wireframe a new application. Starting with sketching, then going from low to high fidelity allows me to figure out the interactions. Given the team has done the research and we do indeed understand the user, it might take me some time to work though things. Since wires are throw away, I really enjoy this process.

I use to use Fireworks, but since it isn't supported on Yosemite, and unlikely to work, maybe at all, on El Capitan, sadly I've moved to Illustrator for most of my wire needs, but will often use InDesign for annotations, Axure for anything prototype. Photohsop now has artboards, which is great for those who design with Photoshop, but it isn't a good wireframing tool. Just because you can, doesn't mean you should. Lots are using Sketch or Affinity Designer, or Macaw.

Use what you are comfortable with, and make sure it can be a part of a workflow to other applications or export well to something you can test with.

This doesn't mean once you are done with the sketching, the actual wireframing can't be done in HTML, Axure, or use something like UXPin or Justinmind. This way wires are already prototypes and can be used to test.

Redesign

Again, this depends on the data. If it just needs a refresh, by all means, skip the wires and make sure the interactions are right with a good prototype. Personally, I would reassess the interactions before a redesign. Are the users still understanding the interactions? Does the current flow still make sense or is there a design pattern that may work better? Just because the business requirements don't think there needs anything new, it is your job to make sure it still follows the UX roadmap you've created.

For this, there is probably more UI and visual work being done and InVisionApp might be the easiest thing for you to use to test with.

Mobile

This totally is dependent on the development team. Meaning, how do they need the assets? Do they need an annotated wire? Do they want the initial low fidelity wires to start from? Or are they more concerned with how the interactions work. Since not all interaction design patterns are solidified in stone yet, it is typically easier for the development team to see the interaction working. Does it slide, pop, fade, explode or something else? Just because you add gestures to your annotated wires does not mean the developers who are creating it get your interaction. 

Depending on your development team, they also might be faster then you when it comes to prototyping the device. I worked with an Android developer recently who would initially tell me, "no, I can't do that", then come back 20 minutes later with it working. Seeing it, then we could say yes, or no that works or doesn't.

There are lots of ways to work quickly for mobile. Pop app, marvel, some of the adobe apps, and lots of other choices, including good old Axure.

So, the trend for me, is whatever makes sense, for the particular use case. How do you work?