General Hack Rewards
These are the rewards you actually select and metaphorically roll dice for.
Security towers generally have local-mission boosts, while objective Hacks generally have Geoscape-level effects. This division isn't absolute, though, with the Geoscape-level effects sometimes occurring on security towers. I'm unsure if it's meant to be an absolute division and its failure to be so is a bug or if it's intentional. In any event, I'm thus lumping them all together into one category.
Many of these Hack effects have a duration I list in a range. This range is an actual random range, but it's decided at mission generation and told to you explicitly when considering the Hack, so you can for example see that a given reward rolled its maximum duration and decide to grab it because of that, which is a nice touch and a bit surprising. Usually games prefer to have that kind of randomization occur such that you won't know the result until after you've already committed yourself.
Also note the duration ticks down when your turn ends. This is particularly important to keep in mind if a Hack effect rolled a duration of 1 turn; grabbing such an effect as the very last action of your turn will ensure you get zero benefit from it, since it immediately ticks down to 0.
Furthermore, many of these rewards are paired, where you'll normally find them together. In such cases, they're clear variations on each other, with one being the easier reward and only benefiting the Hacking soldier, while the other is harder and will instead benefit the entire squad. The reward duration in such cases is always consistent between the two: you won't be getting a 2-turn boost for the Hacker vs a 4-turn boost for the squad. They'll both have the same duration.
Bypass Door Lock
Opens the door without the alert level rising.
This is exclusive to the mission for rescuing a Captured soldier, specifically being attached to the Hack for the Captured soldier's cell door, and is... kind of pointless, because picking up the Captured soldier automatically maxes the alert level anyway, and it's trivial to Hack the door, grab the body, then escape on the next turn.
The rescue mission really needed more polish...
And yes, this means that in those missions you'll always get the door icon here twice at the Hack screen. It's a bit silly, and potentially a bit confusing given normally there's no overlap in possible images between the left 'card' slot and the other two slots.
Integrated Comms
Hacker gains Squadsight for 2-4 turns.
Here is where it really matters that in XCOM 2 maximum firing range on weapons isn't actually a quality specific to weapons: it's not that Sniper Rifles have a higher maximum range than Assault Rifles, it's that Sharpshooters (Who wield Sniper Rifles) have innate Squadsight and Specialists (Who wield Assault Rifles) do not.
Which is to say your Specialist -or SPARK!- will suddenly in fact be able to fire a screen away with their weapon.
I really like the experimentation of Integrated Comms, where XCOM 2 is playing around with Squadsight access and all. It's a cool thing, and I actually hope XCOM 3 is similarly willing to play around with previously-ironclad rules.
Unfortunately, in the here and now Integrated Comms is rarely worth the bother.
First of all, XCOM 2 pressures you to keep the squad pretty close at basically all times, and then enemy pods default to aggressively advancing. In conjunction with time pressure objectives and all, you're usually going to want your Specialist advancing toward the objective whenever possible, and generally this will naturally result in them getting in range of all active enemies, rendering Squadsight moot. You basically need a Reaper aggressively scouting for realistic play to result in a sensible Squadsight opportunity for your Specialist...
... which leads to the second issue: that you usually won't actually want to have your Specialist taking shots at far-off inactive pods, because you want the entire squad ready to pile onto the now-active enemies so you can kill them before they act, and the further the enemy is the more likely it is they'll end up out of your squad's reach but still close enough to advance and attack if you keep moving forward. (Which the time pressure objectives generally push you to do)
In conjunction with the limited duration of Integrated Comms, it's dubious to even grab it early in a mission on the idea that it's better to have it and not need it than to need it and not have it. It may well time out before it has an opportunity to come up, whereas grabbing it is definitely slowing your Hacker down since it requires an action point.
There are edge cases it can be useful in, mind. I've noted before that enemies that don't use Cover also don't use step-out mechanics and so can be rendered impossible to shoot by them standing behind a High Cover object; bizarrely, Squadsight will actually let a soldier shoot such a target anyway so long as someone else can see the enemy in question. This will even crop up semi-regularly via Andromedons, as they like to use High Cover and then when killed leave behind their Shell, which doesn't use Cover. As such, it's possible to end up in a situation where a Specialist can't get a line of fire on a target by moving but can shoot them if backed by Squadsight; if they happen to be in range of a security tower, and it happens to have Integrated Comms, there you go, that's the thing to do.
But this is very much an edge case thing.
Of course, it's normally going to be paired with...

Squad Integrated Comms
Squad gains Squadsight for 2-4 turns.
... the squadwide version.
Squad Integrated Comms is still clunky, but is actually a decent security tower reward. First of all, aggressively scouting with a fast Concealed scout (By which I mostly mean a Reaper) is actually possible to leverage well at that point, with (almost) the entire squad taking potshots at the pod from complete safety. Indeed, the AI copes sufficiently poorly with Squadsight harrying I'd consider this an abusive exploit if it weren't for the context limiting the abuse: the limited duration on Squad Integrated Comms, for one, but more importantly the part where you'll normally be in a mission with time pressure pushing you to keep advancing or else fail the mission. (Thus, you can't afford to stand back and shoot in safety while the AI wigs out)
Second, other classes having Squadsight abruptly gifted to them can find it meaningfully expands their options. A Salvo Grenadier or a SPARK (Via Overdrive) might actually find theirself in the position of wanting to not move (Because they want to spend action points on their Rocket Launcher followed by their primary weapon, for example), but also not quite in range of an enemy you'd like them to shoot; suddenly gifting them Squadsight resolves the problem and lets them do everything you want them to do. Specialists are just genuinely one of the classes that gets little out of Squadsight -among other points, Combat Protocol and Haywire Protocol both function at Squadsight ranges in the first place.
Third, in complex terrain it's a lot more likely that someone can't quite get a line of fire on a target without Squadsight but can do so with Squadsight, simply because everyone is affected and so it doesn't need to be specifically the person doing the Hacking. More metaphorical dice rolling means better odds of it actually happening.
Squad Integrated Comms is still not a great security tower reward, but it's an okay one. Grabbing it just in case it crops up -especially if its duration rolled high- can be a sensible thing to do if you either don't mind risking triggering the security tower's negative effect or have a good enough Hacker they're more or less guaranteed to succeed.
It helps that it has a bit more fun factor than a lot of these rewards, doing something interesting. 'It probably won't help, but would be interesting if it did' is more compelling than 'it probably won't help, and will be boring even if it does help'.
Video Feed
Raises soldier's vision by 2 tiles for 1-3 turns.
This can be a bit confusing in practice, as normally line of sight in XCOM 2 is symmetrical: if you can see an enemy, they can see you, with Concealment as the only qualifier. Video Feed raising line of sight can lead to cases where you spot a pod without activating it, which can be particularly disorienting in War of the Chosen with its target preview functionality -you may end up moving a soldier forward with intent to activate a pod spotted by eg your Reaper, only it doesn't happen because being able to barely see the pod means being out of the pod's vision, whoops!
It also has the hidden benefit of effectively raising the maximum range of your soldiers, as standard firing actions (And a number of special abilities aimed at a specific enemy, too) actually have their maximum range defined first and foremost by the personal sight range of units. Even Sniper Rifles benefit a little, as Video Feed pushes back the point at which Squadsight penalties kick in, and so Squadsight-shooting gets effectively a 4% boost to accuracy. (Not that this matters to this specific Hack reward, but it'll be relevant in a minute)
Unfortunately, Video Feed isn't very useful. XCOM 2's pressure for you to move aggressively, where your units should generally be having each action point of movement go out to just about its maximum distance toward the objective, means that Video Feed's small vision advantage is actually pretty difficult to reliably benefit from; this isn't a game where you can inch forward until you spot your enemies and get the drop on them that way, not in missions that have security towers. Video Feed is thus effectively a small chance to sometimes spot a pod without triggering it, and that's about it... especially since it has a short duration, potentially only lasting the turn the Specialist did the Hack.
Naturally, it's usually paired with...
Squad Video FeedRaises squad vision by 2 tiles for 1-3 turns.
Raises Hacker's Aim by 15 and crit by 25 for 2-4 turns.
Raises Hacker's Dodge by 33 for 2-4 turns.
Raises squad's Dodge by 33 for 2-4 turns.
... its squad-wide self, which is also quite bad.
Hacker immediately becomes Concealed.
It's actually so absurdly good it's a bit difficult to say anything of substance. I guess I could note that it's wasteful to grab it if the squad is still Concealed? But mostly it's just absurdly generally good, and only innate Hack risk/reward mechanics give any depth to the question of whether to grab it or not.
An interesting edge case is Overdrive: if you have a SPARK Overdrive and expend all its action points, then use Distraction, the SPARK will retain its Overdrive status and thus be able to shoot or fire their Heavy Weapon or whatever without ending their turn. Note that it will not reset them to 3 action points, though -in fact, I wouldn't be surprised if Distraction would actually take away action points over 2, though I've not tested that exact possibility.
Hypnography
Lowers all enemy Will by 50% for the rest of the mission.
Another Hack reward I like the idea of but feel doesn't work out so well in practice, Hypnography suffers badly from the fact that the player has very few tools that actually care about enemy Will, and especially that they're all concentrated on Psi Operatives, ie the non-standard class you have to spend a lot of time and resources on unlocking and which doesn't actually benefit from combat experience, reducing the incentives to have them out on missions security towers occur in.
If you are using a Psi Operative in a mission and happen to successfully grab Hypnography, it's nice enough, making Insanity, Void Rift, and Domination all more reliable, with the impact being larger on higher-Will enemies, but honestly I suspect the overwhelming majority of players have never been in a position for Hypnography to have a chance of mattering. It probably should've been cut...
... or Psi Operatives made a core class, as I've harped on often enough.
Naturally, this is another reward that has only the one variant.
Disorient
Disorients all enemies for 2 turns.
This sounds incredible, and it absolutely can be pretty good, but first of all it doesn't affect robots, the Chosen, or anything else immune to Disorientation, and second some abilities work through Disorientation and don't care about its stat penalties; ADVENT Shieldbearers are going to put up their shield regardless, for example. It's really easy to get excited about it and then have it do nothing of actual use. It's further hampered by the fact that the player can, in fact, just use a Flashbang, which is a cheap Item you can build right away and which is pretty useful in its own right; if you're of the opinion Disorientation is great, you're going to be bringing Flashbangs, not relying on the possibility of security towers offering Disorient.
That said, it's good to grab it if you're not confident in your ability to finish a pod that is susceptible, and Codices in particular are made much easier to deal with by Disorienting them, since it shuts off Clone, Teleport, and Psionic Bomb.
As Disorientation also impairs Mobility, it does also slow inactive pod patrols a bit. I'm having trouble thinking of a way that's useful, but hey.
Takes control of a random non-robotic enemy for 2 turns.
This won't actually immediately activate the pod whose member got converted, if you grabbed somebody in an inactive pod. They will trigger once either the pod moves or the unit you took control of acts, though.
Deception is a Hack reward that sounds incredible, and has some potential to be decent, but unfortunately primarily functions as a bit of a trap, where a player activates a pod when they're not in a position to take them out, resulting in the player creating unnecessary risks for their squad.
It's also hampered a bit by the fact that it's restricted to targeting non-robot enemies, as the game trends toward those being the weaker enemies. Grabbing a Trooper following around a Sectopod is not ideal, and if it results in the Sectopod activating when you're not ready it's a really bad deal, probably just resulting in the Trooper dying to little effect while your squad now has to deal with the possibility of a Sectopod lurching out of the shadows and murdering someone. It can grab a Gatekeeper or Andromedon, so it's not like this is an absolute rule, but it's the trend until you're very late in the game, where it's still a concern.
It also doesn't work on anything immune to Mind Control, like the Chosen and ADVENT Generals, so don't be getting excited by the idea of having the Hunter kill people for you for a couple turns.
Particularly horrifying is that Deception does work on hostile VIPs! At least you can potentially walk them away from their car? But yikes, this shouldn't have been allowed to happen.
This probably should've been cut, to be honest. Or had its mechanics significantly reworked, though that would probably require Hacking be approached differently pretty fundamentally to have such an overhaul truly helpful.

Central Command
Takes control of a random robotic enemy for 2 turns.
It's Deception, but for robots.
This makes it on average slightly better for most of a run since robots are stacked toward being nastier than average. It also benefits from it being more plausible that you can predict the target, since robots are a minority; Location Scout, as the simple and extreme example, may allow you to see that there's literally just the one robot on the map, so you can then plan around Central Command grabbing that robot. Among other points, this is actually one of the better ways to get control over a Sectopod, as the odds of success will be better than Haywire Protocol aimed directly at the Sectopod.
That said, one risk of Central Command is that Turrets aren't announced by the Shadow Chamber and maps allowing Turrets overlaps heavily with maps having security towers. It's entirely possible to have the Shadow Chamber announce that Sectopods are the only robot in the mission, get excited when it turns out Central Command generated, grab it, and discover to your dismay a random Superheavy Turret is now uselessly under your control. Surprise!
Overall, like Deception this fits the final game pretty poorly, alas.
Enemy ProtocolPermanently increases the Hacker's Hack rating by 20 points.
Enemy Protocol is an interesting idea whose execution I'm a lot more ambivalent on.
First of all, it suffers hugely in the transition to War of the Chosen: Hacking opportunities are much less consistent in War of the Chosen, as VIP-related missions in the base game always take place on security tower maps, and 40% of them additionally provided an Objective Hack. That's 1-2 guaranteed opportunities per month for the game to roll the Hack reward dice, and Enemy Protocol can show up on both security towers and objective Hacks. In War of the Chosen, your very first VIP-related mission never has any Hacking to do, and even past that point War of the Chosen has added 3 VIP-related mission types, none of which ever allows Hacking to crop up: if I assume every VIP mission type has equal odds of occurring, that still means VIP missions in War of the Chosen only provide a Hacking opportunity 37~% of the time! (That's not really how mission generation works, but the pool system deliberately avoiding repetition doesn't really help; in fact, it means you can't get lucky and roll an extremely long streak of security tower missions!)
That's just VIP missions: Guerilla Ops also notably tilt things away, in that in the base game every Guerrilla Op always took place in the Slum plot type or the Small Town plot type, both of which allow security towers to occur, albeit less consistently than City Centers. War of the Chosen has made it so that 4 out of 9 of the existing Guerrilla Op variants have access to 3 new plot types, none of which allows security towers, and furthermore adds 2 new Guerrilla Op types that also have 3 out of 5 plot types forbid security towers! And neither of these new mission types has a Hack objective, bringing you from 7 out of 9 Guerrilla Op providing 1-2 Hacks and the remaining 2 still potentially providing 1 Hack to only 7 out of 11 Guerrilla Ops providing a guaranteed Hack, with only 5 guaranteed to be able to provide 2 Hack opportunities.
Taken altogether, the base game gives you a pretty steady rate of 2-3 Hacks per month; if you take 8 months, you'll get 16-24-ish opportunities to fish for Hack rewards and by extension for Enemy Protocol to generate: it's not wildly improbable for a given run to have one of their Specialists grab Enemy Protocol 2 or even 3 times, which can easily make it much more feasible to go for normally-risky rewards, including controlling enemy robots -even Sectopods. In War of the Chosen, a run that took 8 months could theoretically get something close to 24 Hack opportunities, but much more likely is that it'll be somewhere below 8. Basically the only reason you have any guaranteed Hacks is because Neutralize VIP missions are forced in late in a run.
"But what about mission types other than Guerrilla Ops and VIP-related missions?" you might be wondering.
Well, the short and technically-wrong answer is that they don't matter. The longer answer is that most Supply Raids never have any Hackables, Retaliation missions never have Hackables, the first three plot missions also don't have any Hackables, Avenger Defenses never have Hackables, Avatar Project Facilities (rather bizarrely) never have any Hackables, and the remaining non-standard missions mostly don't have Hackables. There's only three exceptions: Landed UFOs will have a Hack objective, missions for rescuing Kidnapped soldiers will have a Hack objective, and the second-to-last mission of the game will have a Hack objective. Reminder that Landed UFOs can't spawn until after you've done a regular Avenger Defense at least once, and rescue missions can't occur unless someone gets Kidnapped, which is not terribly difficult to avoid.
Okay, and technically the Lost Towers mission in Shen's Last Gift has a Hackable, but it has fixed reward possibilities, and neither of them is Enemy Protocol.
So yeah, Guerrilla Ops and VIP-related missions are where the vast majority of your Hack opportunities are located, and War of the Chosen significantly reduced the rate at which those provide Hack opportunities, especially from the early-game perspective.
On top of all that, the Fatigue system means that even if your run happens to generate two or even three Enemy Protocol opportunities, you probably didn't bring the same Specialist into every such mission. In fact, you might fail to bring Specialists into any of them!
The reduced opportunities for Hacks also feeds back on itself: not only is Enemy Protocol much rarer to see in War of the Chosen, but a Specialist (Or SPARK, if you take a gamble and get lucky) who successfully gets Enemy Protocol will have fewer opportunities to then leverage their boosted Hack stat, to the point I often don't even bother to try to get it when the opportunity comes up because it's so difficult to care.
In conjunction with War of the Chosen introducing Hack boosts from Covert Ops, I honestly kind of wish Enemy Protocol had been disabled by War of the Chosen.
As for the base game, where Enemy Protocol has pretty good odds of being relevant, I... still am not a fan of the execution.
First of all, I'm not a fan of the artificial incentive to try to have Specialists handle Hacking objectives. Specialists are the overall preferred choice for Hacking objectives thanks to Remote Hacking and their superior Hack score, yes, but Rangers are often the best class at reaching Hack objectives that are positioned so that Remote Hacking isn't very relevant, and 'wasting' an Enemy Protocol roll on a Ranger is unpleasant. SPARKs make things even worse, since they have Remote Hacking but an absolutely atrocious Hack score, where once again it's unfun to feel forced to 'waste' Enemy Protocol rolls on them. As SPARKs pick up Mobility boosts as you tech up, have inherent access to Overdrive, can pick up a move-and-melee attack, and can upgrade Overdrive to let them walk right through walls, SPARKs can easily end up being the unit best-positioned to Hack the objective when you're running low on time and can't wait for the Specialist to get into position. So this is very relevant of an issue!
It doesn't help that, without mods, you won't see what the possible outcomes are until you've already started the Hack, and the Hack-starting animations are all fairly time-consuming: I'm often reluctant to back out and switch to having the Specialist handle things even when it would be more optimal play simply because it wastes so much of my real-life time.
Second, Enemy Protocol inherently leads to jank with Hacking mechanics, in that for Enemy Protocol to be a real reward, the game has to default to Specialists being pretty bad at succeeding at Hacks, a point exacerbated by the decision to have Hack partially tied to level and Gremlin tier and Skullmining: if a Squaddie Specialist is already reliable at Hacking, how is leveling, upgrading their Gremlin, and grabbing Skullmining supposed to be an improvement at Hacking things?
Right away there's something of an anti-teaching moment going on: a still-learning player is liable to try out Specialist Hacking on security towers a few times, get discouraged by how the odds are always dubious or give up when they get too many Hack fails, and by the time their Specialists have enough Hack to be very reliable at grabbing basic security tower rewards, a given player has very possibly long since stopped bothering with security towers. Objective Hacks existing and being very common in the base game helps offset this some by forcing the player to be reminded that the Hack stat is going up, but I suspect there's quite a few players who still more or less completely ignore security towers -especially since it's really easy to overlook they exist if you're not currently Concealed, and the risks involved in Hacking them are much higher if you are Concealed.
But even aside the anti-teaching element, my point is that Enemy Protocol's very existence pressures the design to make Specialists frustratingly bad at actually Hacking things by default, so that stacking three Enemy Protocols onto one Specialist will meaningfully make them better at Hacking tasks. It'd be healthier for the design if Enemy Protocol didn't exist and Hacking was tuned so Specialists were better at it in the first place, such as setting it up so Specialists are more or less guaranteed to get the lesser reward on most Hacks down at Squaddie and leveling and upgrading their Gremlin was primarily about improving their odds of getting the harder Hack reward.
Though to be honest, I feel Hack shouldn't have been based on random pass/fail mechanics in the first place. Among other points, Hacking's design in XCOM 2 has something of a 'winners winning harder' element to it, in that the better you're doing in a mission the easier it is to justify throwing those Hack dice in hopes of getting a benefit; after all, if you're doing great, you can probably readily cope with whatever the fallout for a failed security tower Hack is, whereas if you're already struggling a failed Hack may take things from 'bad, but limping through' to 'total squad wipe'.
Indeed, I consider 'winners win harder' dynamics to be bad design as something of a default. They're useful for purposes like preventing a competitive game's matches from lasting long past the point the outcome is essentially already decided (A classic RTS problem is that often the point at which a player cannot plausibly win occurs several minutes before the game actually declares them defeated, and that's even though the archetypical RTS resource dynamic is fundamentally a winners-win-harder model), but in most contexts they lend themselves to binary difficulty states: either the player hits the ground running and the game is perpetually easy -and thus boring- or the player has a hard time to start and things are perpetually far harder, with no middle road where the game is an interesting challenge, neither too easy nor too difficult. Enemy Protocol itself is an example of such, in that getting lucky with that first Enemy Protocol makes it notably more likely the Hacker will get any later Enemy Protocols you point them at!
... so yeah. I don't like Enemy Protocol, and if XCOM 3 brings back Hacking in a remotely recognizable form, I hope Enemy Protocol is not brought back.
Small Supply Cache
You will gain some Supplies after succeeding in the mission.
This isn't actually carried by the Hacker. I guess the idea is the Hacker discovers the location of a cache that's picked up outside the mission itself? In any event, this is true of all resource rewards you can get at random from Hacking. Which is unintuitive given objective Hacks actually do assign the thing you're grabbing to the Hacker even if it's data or the like.
In any event, Small Supply Cache is okay unless you're up on Legend. A small boost to your Supplies can easily be meaningfully felt until pretty late in a run. It does suffer some from the fact that the better your odds of getting it are the more likely it is you're at the stage where further Supplies are low-value or outright worthless, but not overly-so. The fact that it only rarely shows up on security towers also helps; getting it as a bonus on an objective Hack doesn't suffer from the 'why bother?' problem security tower Hacks often run into.
Up on Legend, it's a lot harder to care. You frequently will be sitting on unspent Supplies for long stretches from quite early in the game; adding some more Supplies to the pile probably doesn't matter.
Of course, it's normally paired with...
Large Supply Cache
You will gain a large amount of Supplies after succeeding in the mission.
... the larger version of itself.
Unfortunately, this gets hit a lot harder by the issue of Hack tuning. Early in a run, Large Supply Cache can be very appreciated of an injection, but if you go for it you're extremely likely to fail; it's honestly better to just go for the smaller but more probable amount. By the time you actually have a high-level Specialist with a tier 3 Gremlin and possibly some other Hack boosts under their belt where you actually expect to succeed in getting Large Supply Cache, you're in the phase of the game where Supplies are rapidly diminishing in value.
Legend of course makes this even worse, with Supplies being more plentiful overall, it taking longer for your troops to level up, and it taking longer to get a hold of the corpses that provide better Gremlins. (No, the Blacksite having an early Mec and the Forge having an early Sectopod doesn't help; you don't loot bodies in those missions)
I basically never bother.

Small Alloy Cache
You will gain some Alien Alloys after succeeding at the mission.
This has more potential to be interesting than Small Supply Cache. First of all, while it's rare, it is possible to get Small Alloy Cache before your first Supply Raid triggers, and thus either get some magnetic weaponry online early or get Plated Armor started researching early. Second, in general you get Alien Alloys at a more controlled pace by default, where it's quite plausible you'll get the Hack, return to the Avenger, and immediately build something you were unable to grab purely due to its Alloy cost, and where you probably would've had to wait several more missions to get it otherwise.
It unfortunately does suffer some from the same issues as the Supply cache Hacks, though, where the more likely you are to get it the more likely it is you won't care. Still, if it shows up early and you get lucky, or even if it shows up in the midgame and you get (less noticeably) lucky, it's generally got a noticeable and interesting impact on your run.
Naturally, it's usually paired with...

Large Alloy Cache
You will gain a larger amount of Alien Alloys after succeeding at the mission.
... its larger self.
Like Large Supply Cache, this suffers badly from the combination of normally being paired with its smaller equivalent and having your initial odds of success very poor. If you care about an injection of Alien Alloys, gambling on the larger amount doesn't really bring anything to the table; it's unlikely you'll be able to say 'getting the smaller amount wouldn't help immediately anyway, let's gamble on the bigger one', especially since the game itself offers zero information on the ranges, and you probably don't remember off the top of your head what Alien Alloy expenses are waiting back at the Avenger.
Similarly, by the time you've got good odds of succeeding at getting Large Alloy Cache, you're probably in the phase of the game where you tend to run out of Elerium Crystals and then sit on unused Alien Alloy piles until you get your next Elerium injection. Notably, there is not an Elerium Crystal Cache, and multiple sources of Elerium Crystals throw in Alien Alloys for 'free', such as one set of Rumors that spawns in the later portion of a run. As such, you basically have to be willing to buy Elerium Crystals at the Black Market fairly aggressively for it to be likely that you're specifically eyeballing Alien Alloys as an important resource in the late game.
The Caches are one case where I think the lesser/greater paired rewards mechanic really doesn't work...


Insight
Current Research has its remaining time halved.
I sort of like the idea here, but I think it suffers a lot from the uncontrolled swing; if you start one of the slower Research projects like Powered Armor and through blind chance almost immediately launch a mission, get an opportunity at Insight, and successfully make the Hack, the effect will be pretty profound, potentially shaving off something like two weeks of research time.
On the other hand, if you get to day 23 out of 25 on a Research and then happen to get Insight, it saves you one day. Yay?
If Insight gave you a little control over its application, like providing an Insight charge you could manually spend at the lab, it would be a mostly-reliable decent little payoff, and there'd be some strategic management questions as far as deciding what's best to spend it on and so on, especially if the mechanics were tweaked so it wasn't so clearly optimal to dump it on the longest Researches. As-is, it's one of the more annoying bits of swingy RNG in the game. The main good news in that regard is, somewhat awkwardly, that it's too rare to be likely to matter; you can go a run of the base game and simply fail to see it at all, and a run of War of the Chosen has even better odds of it simply not showing up.
And yes, this is another case of an unpaired reward with only the one 'tier' but two graphics. This is true of the next four, as well.

Resistance BroadcastBoosts income by 25 Supplies per month permanently.
Another Hack reward I sort of like the idea of but don't feel works out so well in practice; 25 Supplies is a small value, and a run just doesn't do that many months. It's not as egregious a tuning issue as with the Covert Ops, but it does mean you only somewhat notice Resistance Broadcast's impact if you happen to get it in the first or maybe second month, with it becoming increasingly irrelevant the later it generates.
Like yeah sure you might as well have your highest Hack rating soldier go for the Hack in hopes of getting the bonus, but usually you'd rather pick whatever the alternative is.


Satellite Data
Halves scanning times for the next 4 weeks.
In addition to Rumors, this affects Supply Drops and making contact with new regions, though the actual execution is very janky.
First of all, the game insists on only displaying scan time needed in units of whole days, and rounds down this displayed number. A Supply Drop will be displayed as taking 1 day, for example, even though it will actually take a day and a half. As such, it's easy to think you have more spare time than you actually have. This is relatively minor, but still worth mentioning.
Second of all, its effect on Rumors is to halve all current timers on Rumors you haven't scanned at all, and cause new Rumors to generate with half the usual time needed. That is, if you just barely started scanning a Rumor, then a mission happens and you get Satellite Data, the Rumor you started scanning will not be affected by Satellite Data, and conversely a Rumor that generated just before Satellite Data timed out will remain halved forever, even if you don't start scanning it until after Satellite Data timed out.
Third, when considering contacting a new region, it won't modify the predicted time needed to account for Satellite Data, simply telling you whatever the base value for your difficulty is. (Or saying 'instant' if you have Resistance Network active, though of course that will actually be true)
This is particularly frustrating in conjunction with the fourth point: that there's absolutely nothing in the UI to track Satellite Data's benefits. As such, when considering contacting a new region, you won't know how long it will actually take until you actually commit to the contact. Surprisingly, as far as I can tell no mod has been made to correct this, either!
In spite of all the aggravating missteps, this is one of the best Hack rewards, well worth pursuing. Nearly doubling (Remember: travel time is a thing) the amount of work done by the Avenger in terms of scans is very generally great, reducing the pressure from the Avatar Project, bringing in additional resources, making it easier to justify squeezing in a Rumor you'd like to have but might otherwise feel you can't spare the time for, etc. As the Avenger is normally going to be spending the majority of its time scanning, this is all but guaranteed to pay off.
I'm not quite willing to say it is guaranteed; if you get it very late in a run, where Rumors aren't really generating anymore and you're done contacting new regions, it can actually end up doing nothing. (It doesn't affect scanning at an HQ, which is the only scan option that never goes away) That particular scenario is a bit unrealistic, basically requiring you to stall finishing a run, but it can still end up being the case you get it late enough in a run that it technically puts in work but where the work doesn't really matter -you get Supplies from a Rumor faster, but you're sitting on a few hundred Supplies with nothing you're interested in buying as is. That kind of thing.
But mostly, this is incredible. Possibly too incredible, honestly; the other strategically-focused Hack rewards are almost always lackluster by comparison.

Watch ListThe next Contact you initiate will cost 50% less Intel.
As Watch List is a percentage effect, you're actually encouraged to burn it on making a more distant Contact to maximize the Intel savings. As a concrete example, let's say you're in the middle of contacting a region, and your current plan is to build a Radio Relay and then contact another region, followed by contacting a third region. (To reach the Psi Gate, the Forge, or a generic Avatar Project Facility, most likely) By default that plan will, on Commander, cost 40 Intel, then 80 Intel, and if you get Watch List and then carry on with it unchanged it will cost 20 Intel and then 80 Intel, saving you 20 Intel.
If instead you skip building the Radio Relay in the region you're currently contacting, instead contacting the new region and building a Radio Relay in it followed by contacting that third region, Watch List will cause you to spend 40 Intel followed by 40 more Intel, saving 40 Intel in the process. ie clearly superior results.
It's not an essential bit of optimization in most cases, but it's something to keep in mind if you do get Watch List.
Also note that the game won't explicitly tell you that Watch List is active when you're at the Geoscape. If you've got Intel costs for contacting regions memorized, you'll be able to tell because it correctly informs you of what you'll pay right now if you initiate contact, but if you're not paying attention it's easy to overlook that you're paying 40 Intel contacting 2 regions out via Watch List, not paying 40 Intel by contacting 1 region out under normal conditions. Fortunately, any such mistake is liable to be to your benefit.
On non-gameplay notes, Watch List is possibly the strangest of the Hack rewards. The name is a bit opaque, requiring some thinking to arrive at a reasonable conclusion (I'm pretty sure the thought here is that you're grabbing an ADVENT watch list, which is to say a list of people that ADVENT suspects are dissidents or the like, with the reduced Intel cost then representing you having an easier time getting in contact with people friendly to your cause... but I had to look stuff up and think about it, it wasn't just obvious on the face of it), and more to the point about oddness its graphic doesn't mesh with its effects; it uses the same S-looking graphic some bits of the game use to represent Supplies/cash, when Watch List has nothing to do with Supplies/cash.
Did contacting new regions used to cost Supplies instead of Intel, and this graphic not fitting get missed? Did it originally do something else that was/is Supply-related, like reducing the cost of your next Radio Relay? Was it actually a graphic made for an entirely different effect that got cut, and then got reused here even though it's a bit off for this? It's weird.
In any event, Watch List is always good to get. It takes a very long time for Intel to stop being something you can always use more of, and Watch List is affecting your one mandatory Intel expense, so unless you get it really late in a run, where you're done contacting new regions, it's guaranteed to pay off.
Priority Data
Delays currently queued Dark Events by two weeks.
You have no idea how far off a given Dark Event is in the first place and it's absolutely possible to spend months not having Dark Events go through. As such, Priority Data is annoyingly murky as to whether it benefited you at all, and there's no real ability to plan around it; you can't get the Hack and unequivocally decide that it's okay to put off a problematic Dark Event in pursuit of a better reward/less worrying Sitrep/whatever because you got Priority Data.
It's disappointing, because I actually kind of like the idea of a Hack reward giving you more breathing room, but the implementation doesn't really work.
It's an okay thing to go for when convenient, but not anything to get excited about, and a dubious thing to outright gamble on.
And that's the last of the regular Hack rewards. So time to move on to...
Hack Fail Outcomes
This is the stuff that can happen when failing a security tower Hack. More specifically, if a mission has security towers, the game will randomly pick one of the following three 'cards' to be what will happen on failing a security tower Hack. They will appear all the way on the left where an objective Hack would have 'disarm detonator' or the like, so you can see what consequence you're risking with your Hack, which is pretty important given the three consequences are very different in impact.
The prior icon is just a generic 'feedback' icon, where 'feedback' is the game's term for a Hack failure. I'm not sure why it's the term; it sort of makes sense with Skullmining and with Haywire Protocol, but for security tower Hacks it's a bit confusing of a term.
And yes, there's only three forms of feedback for security towers. It's a bit unfortunate, contributing to the shallowness of the Hack system.

Map Alert
Enemies are alerted to your general position, and will attempt to advance toward your squad in the Yellow Alert state.
This is the easiest failure state to ignore, as it doesn't really impact anything meaningful. The game normally tries to have a pod patrolling toward you anyway (And the missions that break from this rule don't have security towers), so it's not like you can sneak through maps in the first place, and the game doesn't attach any mechanics to make it more threatening; the pods won't get to take a shot immediately after activation or anything of the sort. I'm pretty sure the 'downthrottling' behavior that tries to prevent enemies from dogpiling you also overrules it.
As such, if you've already broken squad Concealment, and you check the security towers and see this is the feedback for the mission, you might as well try the Hack if you can spare the action point and all. Even if the only Remote Hacker you have on hand is a SPARK, the fact that they're very likely to fail isn't a disincentive. Indeed, Map Alert can be a benefit, making it easier to get enemies to walk into your squad on their turn and get hit by Overwatch, where they might otherwise inconveniently stand still or patrol away from your squad.
It's a bit unfortunate of a dynamic; this particular form of feedback probably shouldn't have been included in the game.
I do wonder if at some point the mechanics worked such that this was more inherently a meaningful punishment...

Reinforcements
A pod of reinforcements will arrive at the start of the enemy turn, somewhere fairly close to your squad.
The signal flare almost always lands somewhere your squad can currently see. It's not guaranteed, but it's what'll usually happen. If it does land outside current squad vision, it'll still usually land close enough that if you don't deliberately run away from the flare the reinforcements will end up in your squad's sight after their post-activation move. Occasionally the reinforcements won't naturally end up in sight, but this is generally either because a windowless building is involved, or because the flare landed in a low ground area that happened to lack any adequately nearby climb points or ramps, not by virtue of the flare picking a location too far away in a raw distance sense.
In the base game, Reinforcements isn't much of a punishment; it's easy to get the jump on the pod with Overwatch and/or Bladestorm such that the reinforcements get no turn at all, at which point they're free experience and a chance of more loot. Indeed, SPARKs having awful Hack is actually somewhat useful in the base game, as you can point them at the harder reward of a security tower and be basically guaranteed to fail the Hack, netting you the reinforcements that are very possibly more useful than whatever the security tower's intended rewards are. I often actively try to fail such Hacks in the base game.
In War of the Chosen, it works better as a punishment: Bladestorm is unsafe (Unless you get lucky with the rare combo of Bladestorm+Fortress on a Templar... and happen to have them on the mission) due to Purifiers being added to the reinforcement pool, the Fatigue system means triggering reinforcements risks making people Tired who wouldn't have been (Or Shaken, if they were already going to be Tired), and the value of fishing for timed loot is a bit lower overall due to a collage of changes. (eg boosting PCS effectiveness isn't a guaranteed upgrade, making PCSes a little less valuable, your pool of soldiers is wider, making PCSes less consistent a spike in squad performance, Weapon Attachments lose a lot of value as you start looting Chosen weaponry...) For the later game of higher difficulties, there's also the point that some Elite-tier ADVENT units got an HP spike, making it less trivial to stomp the reinforcements before they can act.
That said, even in the base game you shouldn't just mindlessly trigger it the instant you can. Adding a pod when you're already fighting another pod can rapidly go horribly wrong, and security tower missions always means some form of time pressure is involved. With Guerrilla Ops you'll ideally at least get the objective handled first, so you don't end up failing the objective by being slowed by the fighting. If you have Alien Hunters, there's also the risk of triggering Reinforcements and then an Alien Ruler patrols into sight, which is liable to go badly for your squad unless either the Alien Ruler is already nearly dead or you brought a Psi Operative with Stasis; you may wish to skip triggering Reinforcements if you're unsure whether an Alien Ruler can show up in your current mission.

Jammed
+20 Defense and +3 Mobility for all enemies on the map, permanently.
Jammed is by far the most consistently concerning security tower feedback effect. +20 Defense is a fairly large boost, and crucially it breaks a lot of 'magic numbers'. For example, a Colonel Ranger with Blademaster is normally assured a melee hit on the vast majority of enemies, as they arrive at 110 Aim and few enemies have more than 10 Defense. Jammed suddenly knocks your Ranger to an 80% chance to hit against targets with 10 innate Defense, which is not exactly ideal. By a similar token, a Colonel Ranger taking a shot at point-blank with a Shotgun will have a base accuracy of 120, once again assuring a hit on the majority of enemies, but with Jammed you'll end up with only a 90% chance to hit enemies who have a base Defense of 10.
There's a lot of examples like this, where Jammed will turn a guaranteed hit into a risky shot, and of course enemies with Dodge exacerbate this where even if you land the hit you may do less damage than you need thanks to having a miss chance at all. The Mobility boost is less obvious of an issue in the middle of play, as for one thing several melee enemies normally don't show up in missions with security towers and one of the exceptions is generally already fast enough to hit whoever they want (Stun Lancers), but it still means you're more likely to get flanked, makes it harder to retreat far enough if that's a thing you want to do (Such as deciding to just bail in a VIP mission that's gone wrong), and just generally catch you off guard if you've played a lot and have a good sense of where enemies can and can't move and whoops Jammed has changed that on you.
Jammed is in fact problematic enough I personally am reluctant to risk triggering it at all unless it's in circumstances I'm confident it won't matter in. (eg there's only one enemy left, which I can finish with accuracy-skipping tools after the Hack) This is particularly stark in the base game, where Reinforcements tends to be a pinata I actually want and so Jammed is the only security tower feedback effect I treat as A Bad Thing to have happen.
Mind, this is partly commentary on how Hack rewards are mostly lackluster; it's not just that Jammed is dangerous, it's that it's unlikely to be attached to something that's actually worth making that gamble. I'm very much not fond of gambling strategies personally, but even if you're more comfortable with such why bother risking Jammed in pursuit of Squad Reflexes?
Note that Jammed will apply to units generated after you trigger it. (eg regular reinforcements, Chosen summons, Psi Zombies...) Furthermore, it actually applies to the Lost! Including of course that it applies to Lost that spawn in after you trigger it. This can be an unpleasant surprise if you get The Horde on a mission and assume Jammed is harmless because surely Lost don't benefit and whoops they actually do benefit!