Just beam it.
I think one realistic, ish, approach is to use either a thruster or other kind of particle accelerating device to blast the planets surface. In the render i depict this as a ship hovering over the surface, or in an orbit, with the exhaust pointed down whilst the planet rotate underneath it. This does not have to look like this, you may also just collect a bunch of sunlight and concentrate it into a tight beam for the same effect. I just used this ship because i had the 3D model in Houdini.
Using a thruster, such as the Wakefield type relativistic thruster (A particle accelerator) pictured above, may be more to your liking because it produces truly an ungodly amount of radiation. If you simply focus light, you burn stuff. But if you use a thruster to accelerate propellant to relativistic speeds you create Gamma rays and particles similar to Cosmic Rays. Indeed with the setup shown here, the exhaust is best described as a Gamma Ray burst.
This has several upsides. For one, you will Definity delete the planets Ozon layer. With a light beam you have the issue that the atmosphere will absorb a lot of the energy, or reflect it. With particles, whilst the atmosphere still resists it cant really reflect the beam. Furthermore because the beam is so hot, billions of degrees, the instance it hits the Air an effect similar to the explosive shockwave around nuclear bombs may develop. In essence, the beam hitting the air may create something similar to a continuous nuclear detonation.
The exact nature of the event depends on the energy. If we assume the beams total energy to be 100 Mt, it would take around 300 hours to match the energy released by the KT-Extinction event.
Of course in our case, we would match the energy while also irradiating every cm² of the surface to the point the inside of Chernobyl looks like a radiation free zone. Another effect to consider is that since all of this energy cant really escape the planet efficiently, the Atmosphere will heat up to probably a few 100 or 1000 degrees, similar to what happened during KT.
You might ask how plausible such a thruster is. The answer is "meh". Its really about efficiency. Particle accelerators are famous for being heaters that occasionally throw around a proton. But you can get away with more than conventional efficiency because we are building an engine, not a scientific device. And these things dont have to be large. A 10cm long Wakefield accelerator can already approach the speed of light. The reason installations like the LHC are so huge is because they need to get really close to the speed of light, but we are fine with idk 0.999999c. Which is well within the realm of modern tech. (Note, for Particle accelerators people use GeV as the measure of choice, from which you can deduce the velocity).
Naturally, this would be one hell a energy waste. But you also want to sterilize a planet sooooo
EDIT;
Alright because some people take issue with the stated realism, lets try to substantiate this a bit.
I am making, in essence, three cases.
We can build particle accelerators that can achieve velocities of 0.999999c without being the LHC
We can use the particle beam to destroy / Sterilize the planet below
We can supply said accelerators with energy
On 1
That is not up for debate. Within Particle physics the measure of $GeV$, Giga-Electronvolt, is used to describe the power of a given accelerator. To get into some math, the particle velocity is given by
$$v_{p} = c \sqrt{1-\frac{1}{(\frac{J_G}{mc^2}+1)^2}}$$
Where $v_p$ is the velocity, $J_G$ is the energy and $m$ the mass of the particle. $GeV$ relates to Joule by the factor $J = 1.60217 \times 10^{-10}$. If we plug in the mass of a Proton, and assume an accelerator energy of $100 GeV$, the particle velocity works out to $v_p = 0.999957439158c$. Indeed to reach the desired speed of $0.99999c$, we need around $210 GeV$.
Plasma particle accelerators, otherwise Wakefield, are known to reach $2 GeV$ for 2cm track length. Though the record is $4.25 GeV$ for the same length. The Super Proton Synchrotron is currently operating a test article, AWAKE, at $400 GeV$ with a track length of 800 meters. Which incidentally is the length of the Accelerator in the render.
So this is not some sort of super speculative tech, this is what we have. The only question is how much you can scale this up. Because obviously a scientific accelerator is not spewing out the numbers of Protons you would want for a thruster.
On 2
The effects of a particle accelerator beam on the human body is known. One may look at Anatoli Bugorski to get an idea of what a Scientific accelerator can do. In my answer we are talking about a machine that spews out 10000s of times more particles per second. The Beam Anatoli was hit by had an energy of $76 GeV$ and did this;
One metric to look at is the Energy released by the beam per second. This is done using relativistic kinetic energy such that;
$E_{kin} = m_0c^2\left(\sqrt{\frac{1}{1-\frac{v^2}{c^2}}}-1\right)$
If we imagine a beam with a mass flow of say 0.0001 kg/s and a velocity of 0.99999c, the Energy works out to $2.01 \times 10^{15} J$, or roughly $480 \frac{kt}{s}$. Scale this up to 0.001 kg/s and we get 10 times more energy.
Of course, even ~500 kt is not really a big explosion. Its decent but not earth shattering. The magic is the whole $\frac{kt}{s}$ thing. Ideally you would let this run for many weeks or years, and probably not just have one ship doing it. Imagine say 12 ships blasting for 1 year. That will have released the equivalent of 181670 Teratons. 100s of times more than the Asteroid that killed the Dinos. That will, with 100% certainty, obliterate a planets surface and whip out any complex life.
On 3
Now this, is where yall got me. Doing this math is nice and fun until someone asks, where does all of this energy come from ?
At the end of the day, Particle accelerators have garbage efficiencies and if it was so dam easy to do, we would be doing it.
If we stay with the 12 ship example, at the very least we need about you know 24115951226000000 Watt. Otherwise known as twenty-four quadrillion. Also known as like 100 billion times more than the world consumes annually right now.
There are some saving graces. Whilst this is a lot of energy its not a significant portion of the suns total output. Which for all intend is the only possible energy source. If we go with an efficiency of the 1400 Watt/m² on Earth or whatever it is, you only need 17225679 km² of solar panels, also known as the land area of Russia, pretty much down to the km² interestingly enough.
Is this possible ? You know... probably not within the next 20 years. But it is not totally out there. By Sci Fi Standards anyways. So weather or not this is an obstacle depends on OP and their world. If for instance you place the solar panels half as close, you only need the square of the area for the same output. Or roughly the area of India plus a few 100000 km². Power beaming and similar technologies do exist, but are not available at this scale in any form.
Conclusion
Ill die on the hill that from a physics POV, this is entirely doable. From an economical perspective, it isnt in 20 years if OP starts with modern tech, or does not want to expand their timeline slightly.
