Mage, Cook, Archer, Spy: A somewhat unconventional hexblade.

[Corran]

Hexblade 9/ Assassin 3
(Archer build, pact of the blade)

https://kuulpeeps.com/wp-content/uploads/2017/06/wallpaper-assassin-8.jpg


“I don’t know about you, friend”, the ninth replied with a smirk
“But why go through problems when you can go around them
Or walk quietly by them, with them none the wiser?
The ninth—a slender woman in dark leathers, trimming her nails
With her keen daggers as she spoke.

“Stealth, skills, and treasure. These are my job. Do you need in
Where you shouldn’t be? Do you need that one dead, quiet as can be?
Maybe a lock picked, or a purse robbed. I can do that, for a price.
All these things and many more—and unlike the rest, I won’t boast about it later.
Is this not true power?”


TL;DR
Race: Changeling
Pact: Blade
Starting stats: CHA 18, CON 14-15, DEX 14, WIS 10-12, INT 8, STR 8
Note: We might want to give CON an odd score, in case we want resilient con taken at character level 15/19
HP: 87
AC: 16

Background: Criminal (spy variant)
Skills: Arcana, Deception, Stealth, Persuasion, Perception, Acrobatics, Sleight of hand
Tools: Thieves' tools, Disguise kit, Poisoner's kit, Cook's utensils, Gaming set (to taste)
Expertise: Stealth, Deception
Languages: Common, Goblin, Undercommon, Thieves' cant

Progression: Hexblade 5 -> Hexblade 5/ Rogue 2 -> Hexblade 7/ Rogue 2 -> Hexblade 7/ Assassin 3 -> Hexblade 9/ Assassin 3
Further suggested progression: Hexblade 17/ Assassin 3

Feats
4) Sharpshooter
11) Lucky

Invocations
2) Devil's sight
2) Improved pact weapon (taken at level 3 by retraining the choice of the previous level)
5) Thirsting blade
9) Eldritch smite
12) Whispers of the grave

Cantrips
1) Friends
1) Minor Illusion
4) Booming Blade (or prestidigitation)

Spells
*) Hex
*) Invisibility
*) Fly (subject to change, depending on group comp)
*) Hypnotic pattern
*) Tongues
*) Dispel magic
*) Dimension door
*) Galder's speedy courier
*) Banishing smite
*) Scrying






An introduction to the thread and to the build

1) Is this a guide?
No, it's too opinionated to be one, and as a result the focus is extremelly narrow. It's also not a pvp arena, so please refrain from turning it into one. It's just something I made for my own amusement. Initially it was to be just a build which I'd post in LudicSavant's an Eclectic Collection of Fun and Effective builds (https://forums.giantitp.com/showthread.php?583957-An-Eclectic-Collection-of-Fun-and-Effective-Builds), but as I went on and on, I realized that I needed more space and I also decided that it would be better if I just split the content into many different sections. It's theory crafting on some level, as I've not played a hexblade yet, but the build shares many elements with characters I've played in the past, so at the very least I think I have a good idea regarding what I am talking about. The playstyle it intends to offer (better conversely, the playstyle that supports playing the build to full effectiveness) is certainly inspired by what I did with an old 5e character of mine, which was in turn inspired by one (https://www.enworld.org/threads/guide-dealing-death-handbook-of-the-true-assassin.469125/#post-6707781) of GladiusLegis' many guides (which I would not recommend cause it led to some very gimmicky builds; and I am not ashamed to say this, because I honestly think that the author probably thinks the same; I do however recommend checking GladiusLegis' other (https://www.enworld.org/threads/guide-my-word-is-my-sword-the-paladin-guide.474035/) guides (https://www.enworld.org/threads/guide-i-fought-the-law-and-won-the-rogue-guide.488616/)). The poem in the beginning is from PhoenixPhyre, and you can read the full version here (https://forums.giantitp.com/showsinglepost.php?p=23687112&postcount=1).

2) So, why an archer and not an EB cannon?
Because with the right investment, the bow will allow us to spike our damage more effectively.

3) Wait, is this another crit-fishing build?
No. It is not. Unlike a crit-fishing build, we are interested (and it matters) when we'll spike our damage. We'll rely a lot less on the random nature of the d20 (our choices will support this) than a crit-fishing build would, because our aim is to use our nova potential intelligently (yeah, that's right, I find crit-fishing builds to be stupid and annoying and they need to go away; don't take it personally).

4) Why does that matter? Don't I care more about the whole adventuring day's worth in dpr?
Not when having the ability to go nova is involved. If build A does 100 whole less points of damage during the adventuring day compared to build B, but it also managed to take out enemy C three turns earlier than build B did, and enemy C would deal an average of 50 points of damage per round, while the extra dpr from build B would result in taking 100 less damage over the course of the whole day, then build A outperformed build B (and if taking out enemy C three turns earlier happens to be the difference between victory and defeat, I'd say massively so; and this is the real reason as to why we are interested in nova abilities; cause when you can use them effectively, they can push the limit regarding on what it is that you can take on, at a cost you'll have to pay gradually during the rest of the adventuring day; in our case since we have a warlock, until the next short rest).

5) So, are you saying that this build is in fact better than an EB user, or even a sorlock?
I am not saying that. And strictly from a combat perspective, I'd say probably not. But this is a strength we have over those other builds. We paid dearly for acquiring it, so at the very least we need to be aware of it.

6) What ''strength'' is that, exactly?
The ability to take out enemy priority targets and ideally from turn 1.

7) And how do we do that?
By stacking an eldritch smite and a banishing smite on top of one bow attack. We'll need to have two attacks so that we increase the chance of landing a hit, and it would be important to go first on initiative too (cause priority targets tend to cause much trouble on their turn). Ideally we'd like to have the lucky feat too, which will give us a massive boost.

8) Wait a minute. I am fighting an adult Red supported by a horde of kobolds right now, and the Red certainly seems like the priority target to me. The damage you proposed is not nearly enough for taking it out at turn 1.
Well, what I said holds about targets with hp around the region of 122 (ie the average damage of our hit, if you also take into account the effect of banishing smite; this is a relatively safe estimate; we can increase it by taking some extra risk, ideally from when we can be helped by an ally, in which case we almost surely need the safety net of a lucky die, cause ready action allows just for one attack). You can still put your nova to good use against an enemy like a dragon though. Particularly if you reserve your slots for a well timed eldritch smite. In this case, when the dragon is airborne.

9) 122 hp? That sounds like too little at this level.
It's not. Many dangerous enemies have hp around that threshold. Consider yourself as someone who is good at taking out enemies glass cannons (who fight behind more durable allies of theirs). Your range, mobility, the ability to ignore resistances and not count on saves and a few more subtler details, all pile up to a big list of factors that play to your advantage regarding this. The most notable example of such an enemy that I can think of is a lich.

10) Acererak!
No, I am not going down there.

11) What do you mean?
The build is optimized in a more campaign-oriented way rather than a combat-oriented way as a whole. A combat as war mentality will be what will allow this build to shine, rather than playing it in story driven campaigns or dungeon crawls.

12) What on earth are you talking about?
Here is a hypothetical. You've tracked down the hidden lair of the cultists you are hunting down. It's inside a cave system, less than a 2 hour march away (on foot) from lord Bart's castle, where you know that about 100 troops are stationed, and a quick peek revealed probably what is a little more than 2 dozen horses (most likely intended for the elite troops of lord Bart). You have all sorts of clues pointing to that Lord Bart is supporting the cult, and you are afraid that if you attack the cult's headquarters, you wont have enough time to finish the job without being caught between a rock and a hard place (because you know that it is probable for the cultists to have a way of signaling to the castle that they are under attack). A possible course of action could be for you to infiltrate the castle, cause some sort of sabotage (eg burn the stables -I know...- while escaping with one of the horses), and win some time for your allies who will initiate the attack while the castle's forces will be distracted. Or even better, assuming it's worth the risk, infiltrate the castle, kill the lord, impersonate him, and have his forces join the attack against the cult's hideout. There might even be a better plan which does not involve splitting the party, or a plan for which you are not a core piece. But the fact that you have such kind of options available is kind of the point I am trying to make here. Optimizing for combat is one thing. But there are many things you can attempt to do before combat even starts, in order to make it easier or even completely unnecessary.

13) What if the lord has more than 122 hp?
Then get the drop on him (our choices will support this option). This will allow you to push the hp target at around 184. Assassinate comes in to expand our tactical role ever so slightly. Your specialty no longer restricts itself to taking out glass cannons, but also tougher foes, under the right conditions of course (against such enemies our focus is less on immediate action denial, and more on just dealing a hefty amount of damage). Being more self reliant is a welcome side effect that sometimes it might be put to good use.

14) Assassinate? That's 3 levels worth of investment away from your main class.
It's only one level away from cunning action, which we want very much. I'll tell you later why.

15) Even so, getting surprise is hard to do with a whole party following you around. Right?
Yes it is. There are ways to go around that (eg someone uses pass without a trace and/or tell your friends to plan their characters to be as stealthy as they can), but you can always scout a little ahead so that if the opportunity presents to do sth you do it and then you can fall back to your allies who are following you from a safe distance. If you have access to magic items and you are not short on coin (and you are also not a follower of Tiamat), help your allies in getting some items that will boost their stealth. Or even split the party (there, I said it), assuming the risk is worth it.

16) Split the party?! Even if, and that's saying a lot; even if it is a good idea, it's a crappy way to play the game. Hogging the spotlight while everyone else is watching you play and getting bored.
It doesn't have to be this way. Share the action. Either by allowing every now and then someone else to control your character in such instances, or play a mini session with just you and the DM if there's that possibility. You could even skip the whole action sequence of such a plan. The most fun is the inception of the plan anyway, to which everyone can contribute with ideas (and here's the minimum requirement; that the other players like taking the time to think of how to execute such plans, as opposed to preferring hacking and slashing or pure rp). After you settle on a plan and the DM hears it, you might not even have to play it out. The DM could simply say something like this: ''I'll give you a base 80% chance of success, from which I'd subtract an unknown amount (between say, 5% and 15%; the DM might or might not tell you that) based on factors you don't know about, and I'll subtract a random number between 1% and 5% which symbolizes something going wrong due to very bad luck. If you fail by 10% or more you get killed, with the other options being getting captured or success. What do you do?''. I've played like that, and personally, I didn't find myself missing all that much the execution of the plan. Figuring out the plan with the rest of the players (both in what regarded my character but also how that would tie to what the other characters would be doing) was where most of the fun was. In any case, if none of this is for you, you can always skip the 3rd level in rogue (or pick a different subclass). Be warned that at this point, you've got a ton of stuff that have good synergy with assassinate: Hex, branding smite, which at first opportunity we will retrain to banishing smite, eldritch smite, expertise in stealth, a very good range, two attacks (while we only need to strike with one of them to get the big numbers), access to many utility spells can help pave the way towards a successful assassination, and the ability to summon a weapon out of thin air (thanks to pact of the blade) which can prove immensely powerful at the hands of a well played assassin. Plus, it also facilitated by all the social stealth that a changeling can pull off, and also it in turn allows for opportunities for even more rewarding social stealth that the changeling will pull off after the assassination.

17) What do we do when there is not a good target to nova, spell wise?
Whatever it is that other warlocks do. I did restrict my options to the bare minimum I think I'd need, cause I wanted as much out of combat goodness I could get my hands on (as I said, this build is more optimized in a somewhat unconventional way), but you don't have to do that.

18) What do we do when we are out of slots?
As much dpr as we can. Obviously. We are a ranged squishy build. There isn't much else that we can do anyway. And doing damage from range is useful, cause it means we can direct it where we want more easily (this is important, and it's even more important for nova than for dpr).

19) So, EB?
No. I would really prefer to use my remaining invocations for things other than boosting EB. Besides, we have gone some way investing into the bow, that it makes sense to use it for dpr too.

20) How do we do that though ?
The thought behind this is simple. We plan to take devil's sight for a darkvision with a range of 120'. Our aim will be to try and shoot enemies from outside their darkvision's range whenever possible (a high movement rate will help us a lot here). We do that so that we can get advantage with our attacks, which we will combine with sharpshooter. We wont regularly hit as hard as a samurai sharpshooter, but we will be comparable to an EB canon warlock in term of dpr under the right visibility conditions (or against low AC enemies). Being a hexblade will help here too. We are using a weapon, but we are still a caster for the most part at this point. The build may only feature 9 warlock levels, but being 9 levels in and having only dipped 3 levels away from a fullcaster, I would find it very hard to be persuaded to plan for anything other than a hexblade 17/ assassin 3 from a 20 level perspective. We are going to need charisma for our spells, but since we took sharpshooter, we also need a boost to our attacking stat. And as we will see later, we are really pressed for feats. Hex warrior solves this issue.

21) Is this where cunning action comes in?
Yes. Although it's a great feature for any character. But we specifically want it mainly for mobility. We have the ability to shoot from a great range (up to 600'), but also #20 means that we will try to shoot from at least 65' away as often as we can. That means that we will allow for more options for cover to the enemy, than if we were to close in to a distance <65'. We need to be able to move around fast in order to mitigate this potential issue (because although sharpshooter may help with 1/2 and 3/4 cover, it does nothing against total cover). Moreover, as a ranged build, we want to have ways to get away from enemies that managed to engage us. This is especially true for warlocks (with the exception of feylocks), who suffer a high opportunity cost (usually it costs slots; slots that we, as a hexblade archer, have additional ways of utilizing and on which we based a significant amount of what this build can do effectively) when they use one of their various options to deal with enemies that engaged them in melee. Cunning action doesn't come cheap either, but being able to get 65' away or more from an enemy you want to shoot at with advantage from as soon as in turn 1, is helping us play to our plan. Being able to use a BB and disengage/dash (instead of using a precious slot to misty step; hex warrior and pact of the weapon & IPW invocation means that aside for our bow, we can have a melee weapon keying off CHA, so BB just got a boost and so did our melee hit and run plan B) is also another useful contingency. I guess since we will have expertise in stealth, using darkness along with devil's sight, cunning action and BB is another good contingency, but personally I have grown bored of this, and for reasons that have to do with action economy but more generally with how the build intends to function, it wont be all that important to count on this gimmick. Maybe take darkness and keep it until you get dimension door (which is safer than cunning action) for when/if you are caught without your allies around. Last but not least, in some cases we'll try to rely on cunning action for hiding at the end of our turn in order to better protect our hp and concentration.

22) I am digging what you are going for regarding combat, but I am not that interested in what this build has to offer regarding the social pillar of the game. Should I play something else instead?
Definitely. And a gloomstalker ranger is probably what you are looking for in this case. They can easily achieve range/sight advantage (in fact better than we can), they have features that synergize with sharpshooter more than our own (so you can expect a better dpr performance), they have a 1st round nova ability too, and they also have a built in version of our cunning action (or was it the ranger fix?) which means that they don't even need to multiclass. Not to forget that they also get some spell support too (as a half caster), among which you will find pass without a trace, which makes the stealthy approach a viable option for the whole party, even if you have heavies. And they'll be better regarding everything that has to do with exploration. All in all, they are a great subclass, with a similar combat approach to that of our build here. I'll tell you what though. Combining two of those characters would be brutal. I at least, would very much like to have a gloomstalker archer as an ally when playing this build. And if you can think of 3 major reasons as to why that is, when/if you read the whole thread, then congratulations, you have a pretty good idea regarding how the build presented in this thread intends to function.

You do realize that masking this as a dialogue does not make it any less of a wall of text, right?
I know. Shhhh…..

Do you also know that you are talking to yourself right now?
….

[Corran]

Race
There is one race that stands out from all other options for this build: the changeling.
I'd put human (variant) and any race that starts with a +2 to charisma in a tie at the 2nd place, so if you have to (eg AL rules), pick between them depending on if you value faster feat progression or a certain charisma 1/2 feat (eg actor, EA) respectively. But my top pick, and by a lot, is the changeling, so I'd discuss most things with that in mind. The halfelf (that intends to get EA) will get a few honorable mentions here and there, since people tend to like their damage, but unless your group is playing with AL rules, it's not a choice I would recommend. Even with AL rules, I'd probably lean more towards a +2 cha race, so that I can grab the actor feat in place of a charisma bump, or towards a variant human for the extra starting feat.



Changeling

https://imgur.com/ZbT9dsu.jpg

The changeling will allow you to start with a 18 in charisma. This is a big deal for this build. As you will see later, we are really pressed for feats/ASI's (for invocations too). Starting with a 18 in charisma is almost the equivalent of a bonus feat for this build. So this alone almost puts the changeling on par with a variant human. Factor in the extra skill proficiency (and this build can make good use of many skills), and I'd say that the changeling is now slightly ahread of the variant human, who at this point has only a +2 to a tertiary stat to compensate (in this build I'd take the extra skill). But that's only half of it, as far as what the changelings gives us.

The other half comes from the change appearance racial feature. You might have heard how you can combine actor with mask of many faces and possibly with the cantrip friends. This is something that often comes up when discussing warlocks. Well, this combo can be lots of fun, but it's amazing at the hand of how this build intends to play. And it (ie social stealth, so to speak) can have actual synergy with assassinate too. And what the changeling gives you (ie change appearance), is a better version of mask of many faces combined with the actor feat.

To crudely sum up what the changeling gives us, I'd say that it's the equivalent of two very good feats and one invocation (along with 2 extra skills and languages), and we get it from the get go too. The importance of these 2 ''extra feats'' and of that 1 ''extra invocation'' will become clearer once you start to realize how much competition this build has for both its feats and its invocations. I honestly cannot see how any other race can even compare to what the changeling offers.



Halfelf
The halfelf is the second notable option. Comparing it to the changeling, the halfelf has two advantages. It will end up with a +2 to a secondary stat (either DEX or CON), which is not terribly important for a ranged build that is as SAD as this one. But the main thing it has going for it, is that it allows access to the elven accuracy feat. Initially I presented the choice of either going with elven accuracy or with actor if playing a halfelf, depending on if you wanted to boost your dpr or your social stealth more, but with the changeling as an option, there is no reason to go halfelf if you intend to take actor instead of elven accuracy.

Elven accuracy will translate into a dpr boost. Since we are planning to use sharpshooter with advantage, EA will provide a decent dpr boost. This dpr boost will be even better if we enter the fight with unused spells slots, but we wont count too much on that (or at all, at least until we get our 3rd slot; and even then, only just a little), cause we wont sit around saving our slots until we crit. To be fair though, EA will probably get a boost when we get our 3rd, and finally our 4th slot, while at the same time access to mystic arcana. Cause this open up a little more the possibility of crit-fishing without gimping our tactical approach to encounters.

Even without the crit spike, EA will still provide a decent dpr boost most of the time. Is it worth giving up everything that the changeling gives us for the chance to get EA with a halfelf? IMO no. And it's not even close. But this is the most damaging option, so I thought it would be good to give it a mention. Definitely better suited to a more conventional crit fishing build though.

So, if you must...
Race: Halfelf
Pact: Blade
Starting stats: CHA 17, DEX 16 (or 14), CON 14 (or 16), WIS 12, INT 8, STR 8
HP: 87 (or 99)
AC: 16

Background: Criminal (spy variant)
Skills: Arcana, Deception, Stealth, Persuasion, Perception, Acrobatics, Sleight of hand
Tools: Thieves' tools, Disguise kit, Poisoner's kit, Forgery kit, Gaming set (to taste)
Expertise: Stealth, Deception
Languages: Common, Elven, Dwarven, Draconic, Thieves' cant

Progression: Hexblade 9 -> Hexblade 9/ Assassin 3
Further suggested progression: Hexblade 17/ Assassin 3

Feats
4) Sharpshooter
8) Elven accuracy

Invocations
2) Devil's sight (feel free to delay it until level 4)
2) Improved pact weapon (taken at level 3 by retraining the choice of the previous level)
5) Thirsting blade
7) Eldritch smite
12) Tomb of Levistus

Cantrips
1) Minor illusion
1) Mage hand
4) Booming blade

Spells
*) Hex
*) Comprehend languages
*) Invisibility
*) Fly (*subject to change, depending on group composition)
*) Hypnotic pattern
*) Dispel magic
*) Dimension door
*) Galder's speedy courier
*) Banishing smite
*) Scrying


Halfelf variants
I don't think whatever the variants can offer is better for this build than the extra two skills that the classic halfelf provides, but I want to highlight two of the variants.

High elf variant: I noticed a lack of useful cantrips in the warlock's list. This build could make good use of dancing lights (for lighting far away locations in order to snipe enemies) and mending (repair your arrows for starters). Message would be useful too.

Wood elf variant: The extra 5' of movement can be useful for a ranged build with access to cunning action. Even better for someone who intends to shoot from a minimum of 65' away a lot of the time. I mentioned in the beginning how important I consider having a good speed is for this build. It's an option worth considering if you are already going with a halfelf.

[Corran]

Invocations

For good or bad, our first 4 invocations slots are reserved for devil's sight, improved pact weapon, thirsting blade and eldritch smite (which we are going to take in that exact order). So you'll have to wait till character level 12 (warlock 9) before you start having real fun with your invocations. I feel your pain, person who is addicted to misty visions. Your only hope is that you'll find a magic longbow early, so that you can retrain improved pact weapon into something more interesting. But at least for you, person who is addicted to mask of many faces, you may have the option to use a changeling, at least if you don't play with AL rules. There are several invocations that are both flavorful (enough to define your character to a significant extent) and powerful (both in conventional and unconventional ways). So in a way, all 4 invocations that I mentioned in the beginning, are kind of a tax that we imposed on ourselves, by choosing to go with this specific multiclass and split for the reasons that we did. Don't get me wrong, they are the best options once we take into account the actual build, but opportunity cost is what it is. And like with feats, invocations is another area where this opportunity cost presents itself. Let's see though why we need these 4 invocations.

Lets account first for our gimmick (ie assassinate).
Eldritch smite is key here for obvious reasons. Thirsting blade is important too, cause it significantly increases the probability that we will hit with at least 1 attack during a surprise round (and 1 attack is all we need in order to apply all the extra damage from eldritch smite, a smite spell and from sneak attack). Improved pact weapon plays a small part here too, by marginally increasing the chance to hit with at least one attack. Devil's sight is useful too, in a sense that it can allow us to attempt assassinations from further away sometimes, which is a very decent benefit.

Let's account for our dpr.
With so many resources committed to using a longbow already in an effort to hit hard during a surprise round, I thought that the best way to increase our dpr (and we want to do that, so that our build will be as less gimmicky as possible) would be to further increase our longbow's damage output. Thirsting blade is our first priority, but we already want it anyway because it also helps assassinate perform better. Devil's sight (along with the sharpshooter feat) is necessary as it can be an action economy free way to create advantage for ourselves (provided we are mobile enough). Improved pact weapon was also necessary, so that we can increase our attack bonus (as it is what allows us to use charisma instead of dex with a mundane longbow; the +1 to attack is very useful too considering we have sharpshooter, and the +1 to damage is ok), which is particularly important when we will be using a feat that can transform accuracy into a very effective damage boost. Eldritch smite is a potential dpr boost too, given how we'll try to roll with advantage as often as possible (and with some help from hexblade's curse too, assuming we get close enough to activate it). Assuming we have not already used both our slots for something else already.


Regarding the last invocation that the current split will get at character level 12, the are several good options:

Since the halfelf version strikes me as a more combat focused build (at least when compared to the changeling version), I'd take tomb of Levistus with it. We are a squishy build (low AC with low -effective- HP). Yes, we have options to divert attention during our turn (usually by running away, possibly combining it with a booming blade or by hiding/ taking cover), but what happens off turn due to a miscalculation or due something unforeseen like getting surprised (assuming we roll high enough on initiative), being critted, or targeted by something we didn't manage to avoid (like a dragon's breath)? A reaction that can save our squishy butt, surely is a strong addition and something that both versions of this build will need eventually.

With the changeling build, I think I'd go with whispers of the grave. As a squishy ranged nova build, sometimes we might end up killing someone before asking any important questions that we might had of them. Whispers of the grave will try to remedy this. Being a changeling can be of great help here too, since the corpse wont be willing to help an enemy, as per the description of speak with dead. Being a changeling might help even more if you take the appearance of someone you knew that your victim trusted. This last one is not a sure bet, but it's a logical possibility (that you should try to sell your DM on).

https://imgur.com/7BGuYRk.jpg


There are other invocations that I'd really like to get my hands on, especially with such a build. And that's one of the main reasons that I'd continue taking levels as warlock after character level 12 (the other big reasons being access to more spells/mystic arcana and to more slots). But it's a feature that is hard to evaluate objectively, or even evaluate at all. Invocations are a weird mix of flavor and effectiveness, and being forced to choose between the various options is a pleasant headache.

[Corran]

Expertise
Let's pop to the rogue side of the build for a brief moment. The only real choice we have to make is the one regarding expertise. Stealth is the obvious first choice for both builds. For the changeling version, deception is the second obvious choice. I am leaning towards deception for the halfelf version too, though if you really want to focus in combat (always talking about the halfelf version here), you might want acrobatics or perception instead.





A note on assassinate and feats, and the overall importance of the lucky feat
(with a mind to race selection as well)

https://imgur.com/rllOwHG.jpg

So why am I saying that this build is pressed for feats, and thus I am leaning more towards the changeling (or even a vhuman) over the halfelf? Because on an assassin, I'd like to have alert, but even more, I'd like to have lucky. Attempting to surprise an enemy is basically a series of contests.

First there's usually an ability check. Most often it will be your stealth against the enemies' passive perception (though it could be something else, like a your deception against the enemy's passive insight, even a sleight of hand could be involved rarely; these last two concern mostly social stealth, ie your enemy knows you are there, but does not know you are hostile to them). We'll have expertise in stealth and deception, so the chances will almost surely be in our favor. Still, there is a chance to fail.

Then, you have to win initiative. And our build focused on CHA, and not on DEX. Yes, our dex will be in the region of 14-16, but unlike the previous ability contest, we don't have expertise to profit from here. We don't even have proficiency (neither does the enemy, but it was possible for the enemy not to be proficient in perception or insight, as far as the previous contest went). Winning initiative will be the most difficult part of getting the drop on the enemies. Though it's important to remember that every one of these contests carries a risk. Hex might be able to help us lowering the enemy's initiative sometimes. It's not as easy as it sounds (and it may not be as much in accordance to RAW as you might think either), but I'll probably go into that once I start talking about spells.


Last but not least, you need to hit with your attacks. Yes, we'll have advantage (triple advantage with the halfelf), and we only need to hit with one of our two attacks to land both a smite spell and an eldritch smite (remember, we can use eldritch smite only once per turn, and we can apply both the smite spell and eldritch smite on the same attack, so there is little profit from having both attacks hit during an surprise round as opposed to having only one of them hit). Again, the chances here are in our favor, probably massively so. But still, against a high AC enemy you might have something close to a 10% chance to fail at landing at least one attack.

All it takes is to fail at one of these 3 contests, and bye bye surprise and double nova. We need to equip ourselves against this scenario, by increasing our chances of success in as many of these contests as we can. And we'll do that by looking at certain feats. Namely alert and lucky.

On a sidenote, I'd happily trade almost any of the hexblade's features for dark one's own lack (what fiendlocks get at level 6). In fact I'd be happy to do that for a number of features that other warlocks get. It's a sad realization to see that aside from the smite spells (mostly banishing smite) and hex warrior, hexblade offers very little to how our build aims to play. At least hex warrior and the smite spells do a pretty damn good job for us.


This is why alert and lucky are good on an assassin. And you can bet anything that I am taking both of these feats on a single rogue assassin, even on other builds that wouldn't rely as much as we do on spending a feat for sharpshooter and on charisma bumps (both because we are SAD and because we have planned for sharpshooter). Unfortunately, we already have a lot of pressure on our feats, all because of multiclassing and committing to sharpshooter and to a maxed out charisma score.

Not to mention that we are partly a fullcaster (especially if we commit to warlock, as I'd do, after character level 12). We have spells. The best of them will use our concentration. All it takes is a stray arrow from an enemy, and poof, there goes our spell at least 35% of the time. Not to mention that con saves tend to get targeted often enough, and the effects are usually nasty (even something as simple as getting poisoned will hurt our dpr significantly). Resilient con is not something I would prioritize as much as I would feel comfortable with, but at the very least I would want it as a late level pick (even if my plan is to take it as late as at character level 19, ie warlock 16).

Not to mention other feats, like inspiring leader. I don't even want to think about it. I am trying to convince myself that it's not thematically appropriate (well, to be honest, it isn't) and try to completely forget about it. Same with feats like observant, which while not powerhouses, they would be very useful (and fitting) to the type of character we are building.

The point is, that we are really pressed for feats. Every feat we are going to take, all the way up to character level 19, will be a very important addition. So how does that come back to us, always regarding lucky and alert? Simple. With a halfelf, we have no room for either of those, unless we sacrifice either a charisma bump (doubtful) or resilient con (while also delaying a charisma bump most likely). The latter probably being what I'd do in the case of halfelf, which would get me one of alert or lucky at character level 15 (ie warlock 12). That's 4 levels after assassinate kicks in, and 3 levels after banishing smite kicks in. On the changeling, thanks to starting with a 18 in charisma, I'll have room for only one of them (ie alert and lucky), which I'll probably take at character level 11 (ie warlock 8; this means delaying my last charisma bump till cl 15, but so be it). That's just one level after assassinate kicks in, and 1 level before banishing smite kicks in. Much more timely I'd say. All this obviously ties to why I prefer changeling all that much more than halfelf. Cause if you take into account the built in actor/mask of many faces (via the change appearance racial trait) as well, changeling boils down to two extra feats when compared to halfelf. Not to mention getting them from the very beginning, thus we are talking about an unusually fast feat progression. Anyway, back to alert and lucky.

So, at best, we'll have room for only one of them. Which one should I take?

(The importance of the lucky feat) Alert can only help initiative, which granted, will be the most difficult out of the 3 contests. Lucky helps all 3. And the reroll is on average about as good as a +5. So with lucky, we get about the same boost at initiative that alert would give us, while also it can help us with the remaining two contests should we need to. Even assuming you always roll poorly at initiative when you have the drop on an enemy you'll want to smite (which is a completely unreasonable assumption), are you going to use assassinate more than 3 times per given day? Lucky is strictly better than alert for the purposes of assassinate, even taking into account that alert is always on while lucky is not. And since we have invested a lot on our ability to assassinate, I'd go with lucky over alert in a heartbeat. Plus, lucky is very useful on a skill based character anyway, which we are. And we are not a typical skill monkey. This build intends to play a very risky game. And your skill checks will often determine success or failure, and failure could very well translate to character death in some cases. You don't want to fail your stealth check when you are sneaking past the duke's guards in order to assassinate him. You don't want to fail at a deception check that you may have to roll, when impersonating the duke you assassinated in the king's war council. You don't want to fail at your sleight of hand when attempting to steal the treasurer's keys during the subsequent feast. Cause failure at either of these could easily mean death. At the very least it would mean failure of a probably very decent plan. And you would really don't mind if you succeeded at convincing the king, always by impersonating the late duke, that delaying the attack for a week is in his best interests (because it will earn your allies time; time to prepare some city's defenses, or time to figure out where the old king is imprisoned and rescue him, etc). Such skill checks carry a weight far greater than that of any attack roll. Finally, lucky's versatility can be put to good use (eg use it with a save, or with a dispel magic roll, or whatever), if you don't mind spending some of your lucky dice. And as I mentioned in the beginning, having lucky dice will be what will allow us to consider using a ready action for nova delivery, and doing so can be extremely profitable (cause we might want to let one or two quick allies deliver some damage to the enemy we intend on banishing at the very least).

Alert on the other hand, will guard us against surprise, which is important for a squishy ranged build. And it will help our dpr, by allowing us to go before the enemies more often than lucky will (since alert is always on). These are important advantages that alert has over lucky, so it would not be unreasonable to pick it instead. Just know what you are giving up, and don't underestimate the risk that the remaining two contests (other than the initiative contest) involved in assassinations present.

[Corran]

Cantrips
We'll have 3 cantrips at this point. Given that we will prioritize getting at least to warlock 5 before getting any rogue levels, 2 of these cantrips will be taken at character level 1, and the 3rd of them will be taken at character level 4.

Both builds will want to take BB. So that they'll have a melee option, but we want to get the one that has the most synergy with whatever else this build can do. So, in this case, we want BB because it is a melee option that plays nicely with cunning action, which is what we will almost certainly be using if we happen to find ourselves in melee. Now, there is no real rush to get BB. We want to make sure that we'll have it by the time we'll get cunning action, but ideally we want it by character level 5, which is when cantrips power up. Up until character level 5, we can just rely on a normal weapon attack and it wont make much difference that we didn't use BB (in fact, it draws a bit of aggro, which is not good because we are squishy). So I think it's safe to assume that BB will be picked last out of these 3 cantrips, that is at character level 4.

The changeling version will surely want friends, which can combine well with its change appearance racial.

Other than that, I struggle to see anything of particular value. Minor illusion is a safe bet I guess for both builds. We'll be able to combine it with misty visions for one or two whole levels, that is if we take misty visions at level 2, before eventually retraining either at level 3 for iproved pact weapon, or in the case of the halfelf at level 4 for devil's sight (cause there is no big reason to rush devil's sight before we get sharpshooter (since the halfelf already has darkvision), and we get sharpshooter at level 4. But generally speaking, minor illusion is a great cantrip, and the fact that it has no verbal component is particularly useful for a stealthy build like ours.

So, that leaves one open spot in the cantrips of the halfelf (cause the changeling took friends in addition to the minor illusion and BB). Nothing really stands out as too important, but mage hand is a nice cantrip, for which we wont struggle to find uses.

I don't think we have the resources available to make EB worth using. Well, it might be worth using with the right amount of resources invested on it, given how SS does not fair well against high AC opponents (against which we might not be able to get advantage), but I don't think it's worth spreading ourselves that thin for just that benefit.

An offensive cantrip that targets a save might be useful, for when we will suffer disadvantage that we wont be able to cancel, but I think it's probaby something to consider at warlock 10 (character level 13). Though I think I'd probably go with something like prestidigitation, for being able to change the appearance of clothing/equipment that I cannot change from to something else in a quick fashion.






Spells

Banishing smite
This spell will be an upgrade to branding smite as far as our assassinate damage goes (+20 damage on average using a 5th lvl slot), but this is not the reason why I'll pick it up immediately and without any hesitation when I hit warlock 9. The reason I will pick it up immediately, is because this spell, unlike the niche branding smite, has a great effect when used by us. And hexblade archers are among the very few builds that can put this spell to great use (because they get it early, because their slots recharge, but most of all, because they can use it from a great distance and because they can add an eldritch smite on top of it as well). Essentially, it's a nova spell that will add plenty of extra damage to a weapon attack, and if the target you hit with it is reduced to 50 hp or less, you can also banish it (concentration). So its effect cares about the target's hp, rather than saves and thus legendary or magic resistances. And it's a bonus action. This is not a spell you can seriously count on if you and your group are fighting something like the tarrasque, hoping that when you are in the verge of losing you will unleash it and it will turn the tide of battle. Cause in all likelihood it will not. It does not carry that much of a punch. This is the spell you are using in order to take out enemy glass cannons. Enemies with relatively few hit points, who can hurt you much. Most notable example is enemy casters. This is the kind of spell you bring into action when you are fighting something like a lich. And you do it from turn 1 (if you've got lucky, such an initiative roll will be important enough to consider using one of your points, even if it's not a surprise). Ranged dpr builds are generally good against enemy glass cannons. Ranged dpr builds with the ability to go nova are even better in that respect. But ranged dpr builds with access to banishing smite specifically (eg hexblade archers, who can also stack another smite on top of it) are excellent for taking out enemy glass cannons. This is important. It's one thing that not many builds can do, or at least do as effectively. Don't misunderstand. Uniqueness is not the measure here. The fact that it has a very good tactical use is. Just make sure you've got enough range with your bow and remember to stay our of counterspell's range even if you have access to counterspell and the slots to use it yourself (cause it's a bonus action cast).


Hex and RAW
Before I go any further with spell analysis, I have to get this out of the way. Now, one of the things you can do with hex, is to impose disadvantage on one of the hexed enemy's ability checks. Let's say, dexterity checks. Initiative is a dexterity check. And if you've read the section about assassinate, you know that winning initiative against a surprised enemy is necessary in order to crit him. But can we actually use hex to impose disadvantage on an enemy's initiative? Hex has a verbal component, so even if you have the drop on an enemy, and unless the enemy is deaf or there is strong enough environmental noise to cover your verbal casting, initiative is rolled the moment you start casting hex. That's the rules, plain and simple. So the enemy rolls for initative normally, as initative is rolled the moment you start casting the spell, before actually affecting the enemy with it. But. Hex has the beauty that if you are already concentrating on it, say from a previous encounter (and higher level slots increase its duration significantly), then you can move it to a new enemy with a bonus action. That's what the spell's description says. So, presumably, you can do this in complete silence. I am thinking that using hex this way, at the very least would trigger initiative just after the enemy was hexed (if the enemy can sense that he was hexed). Or maybe even not at all. That's most likely a DM call. But either way, the hexed enemy rolls initiative with disadvantage. I am confident enough that this does not go against the RAW, and that's how I would use it, and anything from here on assumes that this is how this will be ruled (so consider hex a definite pick). However, using actions (or a bonus action in this case) outside of initiative, may be frowned upon by your DM who may not allow something like that. You were warned.

Btw, if your DM allows you to transfer hex outside of initiative, it may be in your best interest if the DM rules that the target of your hex is alerted by it and that initiative is immediately rolled. This way, you can use your bonus action to cast banishing smite just before you assassinate, which will naturally drop concentration on hex, but who cares? It already did its job by imposing disadvantage on the enemy's initiative. Don't be greedy by thinking to risk the double crit from banishing smite by relying on only one attack to deliver it, so that you can add 2 extra d6's on the first attack from hex. It's not worth it.


Branding smite
I'll be honest. I wouldn't be looking at this spell if it were not for its synergy with assassinate. Warlocks get few picks, and it's too niche for my taste. That's why I will drop it immediately when I get access to banishing smite. But how does it compare with hex for the purposes of assassinate? Theoretically we could have assassinate as early as character level 8 (ie have a magic longbow by warlock 4, retrain IPW to ES at warlock 5, then take 3 rogue levels). If you can use only one of hex or branding smite against a surprised target (this assumes that you don't already have hex running, or that your DM rules that transferring hex does not trigger initiative), which one do you choose? We have 3rd level slots at this point. That means that branding smite will deliver a little over 7 damage compared to hex. But hex also imposes disadvantage on the enemy's initiative, and it will all likely keep running and make up for the damage loss. For me, it's not a contest, I'd use hex. So, I don't think that branding smite is of any good use to us when we have 3rd level slots. It's also of no use to use when we have 5th level slots (cause we'll have banishing smite at that point). What happens when we can asassinate with 4th level slots (ie hexblade 7-8 / assassin 3). Well, at that point, branding smite will do roughly 15-20 more damage than hex against a surprised enemy, depending on hit chance. Hex still imposes disadvantage on the enemy's initiative, but now it will have a harder time catching up with the extra damage of branding smite (especially if you take into account that we are a sharpshooter build). I'd say that these two levels (ie hexblade 7/ assassin 3 and hexblade 8/ assassin 3) are the only two levels where branding smite deserves a spot in your list of known spells. Assuming you follow that progression of course (the halfelf version of this build does not follow this progression for example). In fact, I might just take branding smite only at hexblade 8/ assassin 3, when I'll also have the lucky feat. At hexblade 7/ assassin 3 I don't have the safety of a reroll, so I might rather use hex instead of branding smite just to increase my chances of winning initiative (my initiative is only at +2 after all).

Note: If your DM rules that an enemy is alerted when you transfer hex on them, and thus initiative is immediately rolled, pick both spells (you still might want to delay branding smite till you get 4th level slots though). This way you don't have to choose, at least any time you surprise an enemy and you have hex already up and running from a previous fight. Just make sure to pick branding smite at the latest opportunity before you have assassinate.


A note about combat spells
There will be encounters where we'll want to spend our slots from turn 1 with a smite spell and an eldritch smite. Usually this will happen when we get the drop against an enemy that is worth reducing their hp by as much as we'll do, or when there is a glass cannon that we want to take out of the fight quickly. Obviously, this type of resource management needs to be done with care. The fact is, that we have additional ways to spend our slots, to those of a typical warlock. Which means that our spells slots face more pressure. Which means, that it wont be worth spending several of our limited known spells for combat spells. We have to restrict our remaining combat spell picks to the absolutely necessary. The fact that our build is a very mobile ranged build that plans to shoot from a great distance, and ideally from a position where the enemies cannot even see us, helps with that. So don't expect to see spells like hellish rebuke, armor of agathys or even something as important as a counterspell (yeah, counterspell's range sucks, but either way, our plan A against casters is to kill them fast; counterspell at best is a plan B if we lose initative and also happen to be within 60' of the caster; and ideally, either of these conditions will be very unlikely, that is if we play and plan our build accordingly). If you carry on with warlock after character level 12 (ie hexblade 9/ assassin 3), you will be only 2 levels away from getting a 3rd slot (at warlock 11). This will allow you to be a little more carefree about your combat spell picks.

- So, what kind of combat spells are the absolutely necessary?
Well, we touched on the smite spells already, and on hex too. A further mention about hex. Aside for how it helps us win initiative against surprised enemies, hex is good because it helps us economize slots, which is what every warlock wants, but we want it even more. So even if it didn't have that synergy with assassinate that it does, I would still pick it.
Is that all regarding combat-only spell picks?
No, we are missing one final touch. Single target damage and nova are good and all, but when you need an AoE, well, you need an AoE. We'll get exactly one such spell. Last but not least, fly will also be discussed at the end of this section as a potential combat buff.


- So, how are we are going to fill our spell list anyway?
Utility. We need it anyway. Granted, some of these ''utility'' spells will have combat applications (eg dispel magic and dimension door), but we select them primarily for their out of combat uses.


Hypnotic pattern
This is the AoE we'll go with. It has a very good range (it matches that of our devil's sight), which means we'll have no problem using it mid fight if necessary from our distant position. It has no verbal component(!), which makes it perfect for when you have the drop on the enemies and there's not a juicy target to assassinate, or when you want to take out multiple foes while making as little noise as possible. And it's a strong spell too.

- Why not fear instead?
Fear gave me some thought. It had some synergy with cunning action, meaning that cone spells need a good starting point, and the mobility from cunning action can help us ensure a good casting point. Plus, as a ranged build But we are fighting from a distance, so it may not always be practical getting as close as we might need in order to ensure good placement of the cone. Getting enemies run away from us can be good, considering we are a ranged build, but we already have cunning action for that. And the fact that we will be trying to shoot while unseen does not play to fear's strengths, as not being seen by the inflicted enemies allows them to reroll their save. The fact that it has a verbal component, while hypnotic pattern does not, is what sealed it for me.

- How about synaptic static?
For most warlocks I might retrain any concentration AoE I would have with synaptic static when I got to this level. Cause even if synaptic static is not necessarily stronger than hypnotic pattern or fear, the fact that it does not require concentration is what would make me take this decision (cause I prefer long duration concentration spells on warlocks generally, allowing me to use more slots with the various non concentration options; and the typical warlock can put to good us many of such options). But we are not a typical warlock. We don't count on running long duration concentration spells all that much, cause banishing smite will interrupt that concentration whenever it will find great use (surprise, glass cannons). So there is not much point prioritizing a non concentration spell to a concentration one, when it comes to our AoE. Moreover, aside for banishing smite which is a definite pick at warlock 9, and considering that at warlock 10 we don't get a new spell weirdly enough (we can retrain an old one to a new 5th level spell though), I want badly 2 more spells aside for banishing smite (more on that later), so that alone pushes considering selecting synaptic static all the way to warlock 11 (ie character level 14). That is if we continue our progression as a warlock. At that point, I will probably choose to include synaptic static anyway (probably by retraining hypnotic pattern, if charm immunity is common enough among the enemies against who I'd usually want to use an AoE).


Dispel magic
Dispel magic can be used during combat (if you have the slots for it), but we mainly want it for non-combat encounters. Ie magic traps. You'll often want to be in places where you shouldn't be, so your are bound to come across some of those. You might want to pair this spell with the eldritch sight invocation (for example by retraining IPW if you have a magic longbow by then).

- Why not counterspell?
As mentioned previously, counterspell uses slots during combat, so slots we may very well not have. It also has a poor range (at least compared to our longbow's range, or at the very least to devil's sight). And last but not least, counterspell gets used against casters, but you already have a better plan against casters, and counterspell is not that much of a plan B either because of the restrictions already mentioned. So pass.


Dimension door
Teleports are generally good for a ranged squishy build, but once again, it's the out of combat use that intrigues us here. Dimension door can help you get into places. It's not the only way obviously, many spells can help you in that respect (invisibility, fly, spider climb, gaseous form, etc), and some of them do that better. None of these spells will beat dimension door for escape though, at least most of the time. Just make sure you are out of counterspell's range if enemy casters are around (move + cunning action dash/disengage) when using it to escape. Or use dimension door and cunning action hide if you want to stick around but at the same time you want for the enemies to lose you.

- What about misty step?
I'd definitely want it before I get dimension door, cause cunning action aside, I do need a teleport for when being grappled or restrained (that is if I don't want to risk a contest where I'll use my acrobatics). But it doesn't compare with the utility that dimension door provides (no need to see where you are going, far greater range, and the ability to bring someone with you). It's action economy advantage is still something, but then again, you can always use dimension door with cunning action to hide if action economy is important. I can understand why someone might stick with misty step, even after getting cunning action but especially dimension door, since it allows you to use your action to attack when you want to escape a grapple or a restraining effect. But is spending a known spell worth it for such an extreme scenario? I'd say no.


Invisibility
Let's start talking about some serious utility, shall we? There is not, and will never be, a good reason not to take this spell on a build like this. Almost everything that this build is good at doing, relies at least on some level on stealth. And invisibility can allow you to roll stealth checks when you otherwise wouldn't be able to. That's it. The fact that it scales well with your pact magic slots (finally allowing you to take up to 3 other creatures with you), which also happen to recharge on a short rest, is good. And it can be something that you might be able to put it to very good use (depending on group composition). But even then, that's just the side benefit. The important thing here is what I said in the beginning.

Did you infiltrate the enemy camp disguised as one of their soldiers, with the plan to steal their battle plans during the night, but you find that the entrance to the war tent is guarded? It's better to rely on a stealth check than on hoping to quickly and silently incapacitate the guards (via assassinate or hypnotic pattern -doubtful- or whatever). That's where invisibility comes in, allowing you to roll for stealth when initially it wouldn't be possible (at least without causing a distraction that might for a split second capture -some of- the guards' attention; which may always backfire).

Do you want to follow the henchman you spared in the hope of following him to his cult's secret base? Invisibility. And if you are lucky enough and there is not an unusual amount of doors, you might even follow him inside and overhear a useful thing or two.

Even as a precaution. Do you suspect that your party is being headed to a trap while on the road? Use invisibility and even take with you one or two allies (the really stealthy ones), moving a few hundred feat ahead of the rest of the party, just in case you notice the trap/ambush/whatever (just keep dimension door handy just in case).

Sometimes it might just be necessary in order to set up an assassination. Other times it will be just to help you steal something. This character is best at attempting some very risky things. And invisibility is one of your tools that will reduce this risk. And other times it will be just what you'll use to give yourself the edge you need in order to actually succeed at these things. It will be, and will remain, one of our most important spells.


Scrying
This is what I am using when I want to spy on someone but I am lazy about it. I am kidding. You want this spell because you can't be everywhere at once. And because it will usually be safer than physically spying upon others, if the latter is even possible at all. So you might be using this in order to overhear the bad guys plotting. Or you may using it to learn more about the routine of a target you want to take out. Or even for familiarizing yourself with someone whom you plan to impersonate. Whatever case, it will provide you with information, but the important thing is that it does so 'now' (assuming the target fails their save of course), rather than say, travelling for a week to physically spy on someone. An important thing to notice is that it even allows you to spy on a place, instead of a person. One reason why this can be important, is because scrying a location does not suffer from the same limitation that scrying a person does, being on the same plane of existence that is. Scrying can often create a false sense of security though. Don't think that it's completely safe, especially if your DM is crafty. First of all, you might get noticed. Worse, you might get played. Also know that there are harmful things that can affect you by just laying eyes on something, regardless of distance (and don't be too surprised if your DM deviates from how 'distance' is defined in the dictionary -lol- for what we are discussing). Last but not least, be careful if you are using this spell to scry a location (which you must have seen before; do nightmares count?) on another plane of existence. If you end up scrying in the abyss, it will probably not end well for you. Caution and moderation when using this spell, as with everything.


https://imgur.com/IRzdPet.jpg

So you shined your crystal ball and you set it on the small wooden table around which your allies have also taken their seats. You all know (from what you managed to uncover earlier this week) that at this time tonight, the dwarven lords will have their meeting. It's about time you started casting your scrying spell. But first, you take out that lock of hair from the pouch you set at the table next to the crystal ball. Hair from one of the dwarven lords! You paid good coin for it, but so be it. It will be worth it a thousandfold when you uncover the location of the lords' secret treasury. And that's bound to happen tonight. You've all worked hard for it, left nothing to chance and you know you'll be successful. You finish casting your spell, and as expected, connection is established and the signal is good. The last of the dwarven lords are taking their seats, everyone else of lesser station has finished exiting the room, and the mumbling dies down. Their meeting is just about to begin. Everyone around the wooden table smiles, as the conversation between the lords is about to begin. And so it does. One of the lords stands up and starts talking to the rest..... in dwarven! Everyone sitting around the small wooden table seems confused. You all start looking at each other nervously. And that's when you realize that you indeed overlooked a simple minor detail. None of you can speak dwarven...


Tongues (or comprehend languages)
Spying on people for information requires at the very least being able to understand what they are saying (sometimes what they have written down too). This is where tongues (or comprehend languages) comes in.

Tongues and RAW: It used to be in previous editions, that this spell actually allowed you to speak a language (or so the internet says). In 5e though, there is some ambiguity surrounding this spell. It does not specify that you are actually speaking the language, or that you are perceived as if you were actually speaking the language. It only says that while affected by this spell, you are understood by anyone who can speak at least one language. It says nothing about what the other person actually hear. As most things, this will likely be a DM's call. And it's an important one for you (at least if you are playing the changeling version, or if you have taken actor and mask of many faces, or if you have spared a few points for INT and have expertise in the disguise kit).

If the DM rules that tongues allows you to actually speak a language (or at the very least that you are perceived as speaking one), then it's what you pick (just be careful when talking in front of members of a different race). Obviously an easy choice, since this spell will help us tremendously when trying to impersonate someone whose native tongue is not one we naturally speak. If the DM though rules the other way, then drop whatever tool proficiency your background gave you (as a changeling we got cook's utensils and gaming set), and replace it with extra languages, and pick your languages carefully. And then pick one of tongues and comprehend languages (you still need one anyway), depending on if you value easier communication over being able to understand writing in a language you don't speak, or vice versa.

So, you finally got your hands on some good information that you also need to share (with npc allies, your boss, or some client). How will we do that?


Dream (to be taken at warlock 10)
The power of long range communications will be only as strong as you'll make it to be, but the sky is your limit. If sending is a pair of two brief text messages (up to 25 words each), then dream is a skype call with more than enough duration for what you'll typically need. You don't use this spell just for conveying information, you also using it from acquiring it (from your various contacts; you start with one, no reason not to develop more over time). Or for setting up complicated plan with distant allies in advance. Bottom line, there are many things you'll be able to accomplish with a skype call in a world that is partly medieval (all right, maybe this last one is an exaggeration, but I hope my point comes across well). But besides effectiveness, this spell is a roleplaying goldmine. You don't get to control just what your character is saying or doing (with only your imagination -and not rolls- limiting you for the latter), you get to control the whole scenery of that ''dreamspace''.


https://imgur.com/8j1nLaZ.jpg

It has two major downsides. The first one is that you wont be able to use it in order to get in contact with anything that does not sleep (eg elves, undead, constructs, coffeelocks, etc). The second weakness is that you can only use it to establish contact with the target when it will be sleeping. So you wont be able to count on this spell to communicate something urgent, like ''it's a trap, abort!''. You still need something like sending for that, and it is a sad fact that hexblades don't get access to sending.

- I've read that you can use dream to kill enemies. Is that true?
It's possible. I wouldn't go there though. At best, you use dream like that in an effort to catch your DM off guard, regarding their mastery of the system. And that's even assuming that your DM wants to stick strictly to what exists in the books (which of course they don't have to). This will result either in restricting the DM's choices regarding the bad guy they are using, or in turning the game world into a joke. At worst, the DM advances their system mastery (or pull stuff out of their butt), and now it's a game that two can play. And guess who'll be better at it. I am not saying ''I wont use dream this way because it will backfire''. I am saying ''I wont use dream this way cause nothing good will come out of it and at best case scenario it will backfire''. But you do you.


Galder's speedy courier
This is a beautiful spell (from 'Lost Laboratory of Kwalish'). Aside for any communication needs it can fulfill (and we'll get back to this), it can be used to transport objects (no larger than 3x3x3; an ally with enlarge/reduce can help here). It does so in relatively good safety, so at worst it can save you time. At best though it's a back up that you can use if you are in trouble, and doing so can at the very least be a negotiation chip if the trouble proves to be too much. You can also use it to expand your communication abilities, as aside for not having to meet dream's requirements, with a bit of pre planning it can be used when you need to transmit something urgently. For example, putting using the spell to send a red ball could mean ''it's a trap, abort!'', a green ball could mean ''proceed as planned'', a blue ball could mean ''proceed with plan B'', etc. There are a few shenanigans that come to mind when reading this spell (which may or may not work), but all in all I think that the spell is useful and flavorful enough before we even try to cheese it.


- What about illusory script? Does it have any uses at all?
Yes, it does. Before you any of your fancy communication tools, you'll have to rely on written messages for when you want to pass information without being present or to someone who is far away. You will either use drop spots or trusted(?) couriers (be they humanoid, a trained pigeon, or whatever else). Once the message leaves your hands, you cannot guarantee what will happen. It could never reach its destination, or worse, it could be worse if it falls into the wrong hands. Illusory script will offer some protection for when the latter happens. Ideally you use it to both convey the right information to the right people and the wrong information to your enemies. Or to incriminate someone. Or to bluff someone (''that's the orders for the prisoner's transfer''). With enough imagination, it will have plenty of good uses regarding the social pillar of the campaign, at least during the low levels. Feel free to retrain it once you get access to either Galder's speedy courier or dream though. It wont lose all its value, it never will, but truth is that it can only do well some very specific things, and at higher level we can put this slot to better use by retraining it to a higher level spell. At least that's what I'd do.


- What about contact other plane? Isn't that a good spell for the purposes of acquiring information?
It could be. It's a very DM dependent spell. It could range from utterly useless to very strong. I definitely like it. But not more than scrying or dream. That's why I would start considering it at warlock 11. If you prefer it, consider picking it earlier though. If you know beforehand that your DM will help you make a very good use out of this spell, consider even starting as a rogue for the INT save proficiency. Lucky rerolls will be of use too in that case.


Fly
Time to pick our last known spell. Now, fly is not a definite pick. There are several other spells I might want instead. Like suggestion or charm person/monster, or something like a hold person or banishment, perhaps even hallucinatory terrain. How about a spider climb, or probably even better a gaseous form. This could have some uses when other spells fail. But before I look at any of these options (and more), I would look at fly. Not for the utility it can offer, but for combat. Because fly, if used as part of an overall strategy, can be very effective for the right group. That wont be the case for every group, but if it is for yours, then pick it, and don't hesitate to use it as part of your group's main tactics. And due to how pact magic works, warlocks have a far smaller opportunity cost rearding using fly as a multi-target combat buff. I will try to put this into context with an example.
The person I am responding to would be playing a single class sorcerer (of undecided origin), so a big part of my reply to them is about their build. But then I jump into some suggestions about group tactics, and as you will see, the warlock's fly spell will be vital for the suggested plan to work. I'll edit the post a bit, to trim out any talk irrelevant to our topic, but if you are interested in the whole discussion, here (https://forums.giantitp.com/showthread.php?595061-Tips-for-Sorcerer-pures-no-multi-classing-plz&highlight=group+tactics) is the thread in question.


So this is what my Dm text me about creating a character
Stat array is 18 16 14 12 10 8
Starting at Level 11, get 1 Starting feat, 1 uncommon magic item.
Plus chose 2 from this list
Learn a language
Profency in 1 weapon
Profency in 1 armor
Profency in a skill
Profency in a tool set

He also has a rule you can't pick a class someone else is already playing. The party is
1. Human raised by minatars Samurai build as a hit and run. She bull rush a enemy hits them and keep going.
2. A bard dont know much else he join when I left for the summer to work.
3. A dragonborn warlock I cant remember how hes built.

So what do yall say I should make?


(snip)

You party composition is rather squishy and you are a bit light on the (single target) dpr department too.
(snip)

Let me expand a bit more on party composition and what it means for your choices. You are 4 characters that don't really want to get hit. Especially the spellcasters who can and will need to concentrate on spells. Bard, sorcerer, warlock, all of you don't want to get hit. And the fighter wants to play hit and run too (which strangely I find a good idea), but either way, fighters don't have the raw defenses or stickiness to tank for everyone in the party. What does this mean for you? That you want to pick up things that:
1) Increase your mobility (or better, depend on flying speed)
2) Allow you to escape close combat
3) Buff your defenses (pick up cheap things for that, don't go overboard, you are not a tank)
4) Buff (protect) allies in an efficient way

(snip)

For example, consider this. The warlock concentrates on fly (upcasting it naturally, to affect 3 pc's). So, the warlock uses fly on them, on the bard and on you, the sorcerer. Everyone is up in the air (avoiding melee enemies), except for the fighter who is down on the ground trying to use hit and run against enemies (hopefully outrunning some of the melee's). How do you build on that? You use greater invisibility. But on whom? First of all on the warlock, so that they don't easily lose concentration on the fly spell which protects the 3 casters from melee enemies. But also because it will improve their dpr (and you warlock is the actual ranged dpr of the team, so that's a big plus). But you also use it on the fighter, who is the only one left one the ground, so most likely the one taking the majority of hits. So now, the warlock, the bard and the sorcerer are all up in the air (with the warlock invisible), and the fighter is on the ground invisible, exchanging hits with the enemy. That will be a good way to go about winning some tough fights. And when you hit level 14 and you get wings, then the figher can be on the air too, which means you could potentially target someone else instead of the fighter with greater invisibility (you still want to buff the warlock, both because dpr but also because they are the ones keeping you up in the air, so you want them avoid getting hit and having their concentraton tested). For example, at level 14 you could buff the bard, who will be probably concentrating on sth as or more important as fly.

(snip)

[Corran]

Progression

Do I start as a warlock or as a rogue?
If you start as a rogue, you'll get proficiency in DEX saving throws (damage, restrain) and in INT saving throws (a fair few nasty effects; synergy with contact other plane), and you'll end up with an extra skill proficiency. If you start as a warlock, you'll get proficiency in WIS saving throws (mostly stuff that shut you down) and in CHA saving throws (a fair few nasty effects), and you'll get quicker access to extra attack and to 3rd level spells and slots (both of which are important). For me, it's mainly a question of if I prefer the extra skill proficiency (which I admit that I would really want), or proficiency in WIS saving throws. And I am leaning to go with the safety of being proficient in WIS saving throws. If for some reason I ended up playing this build with a race that does not grant even a single extra skill proficiency, I might have to give this some more thought.

When do I take the first 2 rogue levels?
I'd take 2 rogue levels after warlock 5, that is after getting my extra attack (ie thirsting blade) and access to 3rd level spells. That's because I don't want to delay my dpr, but also I would like to take some pressure off my slots (now that I've got some good options for their usage; even hex got upgraded with 3rd lvl slots) by relying less on misty step for the much necessary mobility. So yeah, I'd definitely want to be warlock 5/ rogue 2 at character level 7.

When do I take my 3rd rogue level?
Then it's a question of when I want to take the 3rd and final level in rogue. I want to do that after I have eldritch smite. Devil's sight and thirsting blade take priority over eldritch smite, so when to take eldritch smite is a question of whether we can get rid of IPW or not, which in turn is a question of whether we have found a magic longbow. In short, if we don't find a magic longbow, we take eldritch smite at warlock 7, thus I'd like to get my 3rd rogue level just after warlock 7, for a progression of warlock 5 -> warlock 5/ rogue 2 -> warlock 7/ rogue 2 -> warlock 7/ rogue 3 -> warlock 9/ rogue 3. But if I do get my hands on a magic longbow, I would be willing to get the 3rd rogue level at any time after warlock 5/ rogue 2, assuming of course I retrained IPW to ES already. So I could have assassinate as early as warlock 5/ rogue 3 in some cases.

Note: With a halfelf, I'd follow a different progression. Since the reason we picked halfelf is to get access to EA, it makes sense not to delay it. And I want to take EA after I have sharpshooter, and at about the time I will have eldritch smite. For these reasons, I'd rush hexblade 8 before taking any rogue levels. And being one level away from banishing smite, I think I'd also advance to warlock 9 before taking any rogue levels too. So probably we are looking at a progression of hexblade 9 -> hexblade 9/ assassin 3.

What if I am playing with AL rules?
With a variant human I would follow the progression I mention in the beginning of this section (ie the one that the changeling is using). With a +2 CHA race that intends to take the actor feat (which I'd want at lvl 4), I'd probably follow the halfelf's progression that I mentioned just above, just so I can get faster access to either sharpshooter or lucky (depending on if I'd want to focus more on my dpr or on my nova and skills respectively).




Life after level 12
We could take levels in another class. Two levels in druid would give us wildshape, which would be a useful addition (maybe alternative is a bit more accurate) to our scouting/infiltration abilities. A third level in druid would give us access to pass without a trace too, which would prove useful in our hands for obvious reasons. A level in fighter for the archery fighting style would be a decent boost to our dpr as well. Action surge can never be bad for a character with a strong at will. Maybe a third level in fighter would be tempting if you like something like this (https://forums.giantitp.com/showthread.php?601733-Hilarious-Idea-Regarding-New-UA-Wallhack-Warlock-Assassin&highlight=ghostly+gaze+fighter) (not sure if it made the cut from UA). Three sorcerer levels wouldn't be bad either. This character wouldn't struggle to find uses for something like a subtle suggestion, and access to the detect thoughts spell would be very appreciated too. More rogue levels wouldn't hurt either. Faster feat progression, with an extra feat near the end, reliable talent as a capstone, 2 extra slots of expertise, more sneak attack and other goodies (this is a particularly tempting option).

Personally, I'd go with more warlock levels. Granted, I'd value a lot of the things that levels in some of these other classes would get me, but I don't think I need desperately any of these abilities for my build to click. So, I think it's best to judge features in isolation (mostly), in which case I am tempted with advancing my spellcasting (especially since there is enough room for 9th level spells for a level 20 build). But it's not just spellcasting (extra spells and access to mystic arcana). I would also want the two extra invocations. But even more, the extra slots, which would allow me a lot more flexibility in between short rests. And since every last feat/ASI counts, even without thinking too much about the choices I would make, I don't see many dead levels in the late warlock levels either. Well, maybe warlock 14 will be a little lackluster, given the small range of our curse. Evidently, this is not something I have put much thought into, and it's possible that at some time in the future I might revisit this section. It will probably be from the perspective of a hexblade 17/ assassin 3 build though.

[Corran]

Closing words
I am not sure what kind of discussion such a thread can spark (if any). But as I said in the beginning, this is something I made primarily for my own amusement. I probably should have taken one or two extra days to give it a second look, possibly re write some parts so that they are easier to read at the very least. But I think most of what I wanted to include is there (even if the words are in the wrong order sometimes), so that makes it good to go as far as I am concerned. If you had fun reading through any of it, or if it gave you and idea or two, that can't be bad either.

[Corran]

Just in case I need more room (probably reserved if I decide to expand on the cooking side of the build). Upon a closer look, turns out that ingested poisons are not as spectacular as I thought they were, so ''cooking'' didn't get the attention the title hinted at. Proficiency with a cook's utensils can still be useful, but not any more useful than any other sort of tool proficiency you might expect of someone (whom you plan to impersonate) working and/or living in a rich household/castle to have.

================================================== ===

Inner (after)thoughts


Think more about clothing. Preparation, but also when you need to do it quickly. Ideally without having to rely on illusion magic for it, if it can be avoided (heavy armor causes issues, so might have to rely on illusions for this one if it's really needed). List of things to look at:
1) Disguise self/ mask of many faces
2) Prestidigitation and some other cantrip (maybe?)
3) Magic/wondrous items (shiftweave, cloak of many fashions, glamoured studded leather, hat of disguise, glamerweave)
4) Weaver's tools proficiency (preparation, discretion). Hmm, mage tailor sniper spy?
5) RL ideas (magician tricks - multilayer clothing, double face clothes)

Think more about the possible value of int checks. Might have to drop some points from con (most likely) to aim for a better int score. Investigation is also probably worth getting. Athletics might also be necessary if it's too taxing replicating with slots, but skill proficiencies dont grow on trees.

Changeling was errata'd to not be able to start with a +3 to charisma. Trouble. Probably still a little better when compared to the variant human and the halfelf, but not by much anymore (at least under point buy). Feat progression takes a big hit, though it will probably just translate to a worse dpr (still bad).

Languages. Campaign dependent, but take a closer look at:
https://forums.giantitp.com/showthread.php?622776-Best-Languages (sphinx - how I wish!, celestial - check the lore)
https://www.reddit.com/r/dndnext/comments/hha9iy/most_spoken_players_handbook_languages_among/
Narrow it down and put it in context.

Poisons. Contact could be applied on ammunition (at least in theory), so that opens up a few possibilities. Unfortunately the poisons that play to our BS combat approach (with party) are the most expensive ones (but they are still one of the very few things that can improve it). Might be worth looking at poisons more for their situational value, or even as a way to save us (allies included) a spell slot (resources in general). Resistances and immunities in abundance, so use with care. Might give us a very necessary edge in assassinations, especially before we get a 3rd slot (when resources are fewer, chances of escape compete even harder against the chance to succeed; big room for grave error). Think harder about ingested poisons!

[JackPhoenix]

You know, the images almost made me report the thread out of habit without checking what's up and who is it from....

[Corran]

You know, the images almost made me report the thread out of habit without checking what's up and who is it from....
Good lucky. :p

[saucerhead]

Thanks for this, Corran. I found it an interesting read.