To define a design space you need to first design what space you have. For many designs in PF2, this often is done by the negative space of what you cannot do with a skill feat, when it is okay to change those things, and when doing so isn’t disruptive to the design or concepts present in the game.
First, let’s start with the basic criteria for a skill feat:
- It interacts with a skill or skill action.
- It is below the power level of a Class Feat
- It can be taken by many different types of characters
These 3 things are truly all the things that you need to make a skill feat. Truthfully, if you have 1 and 3, you have a skill feat, number 2 is only a guideline for a power ceiling.
Second, let’s talk about things to avoid in PF2 skill feat design:
- Do not layer multiple prerequisites for skill feats across many levels. – The reason for this one is because feat chains, in the grand scheme of character development, actually feel like “feat taxes” to characters after a certain point. If you are not keeping up with your feat chain, then you feel an “opportunity cost” loss, which means you either have all your feats spoken for (rigid character design) or you feel less impactful because you didn’t continue your feat chain.
- Avoid math fixers – Math fixers should be used sparingly, and most often, should be done as a means to resolve a problem with character concepts not working due to the math, instead of boosting existing character concepts that already do succeed at the math.
- Avoid Skill Action Swapping – Skill swapping is much the same as the math fixer problem. It serves to make a character who would otherwise be “bad” at something, good at that something, usually because they would otherwise not be able to interact with the concept because they do not have the resources to participate otherwise.
Lastly, let’s evaluate some skill feats that avoid these problems, but also some skill feats that break these rules, so we can see when to use these as guidelines to break and when to stick to avoiding them.
Bon Mot – This is a bespoke action that has no prerequisites, isn’t a math fixer, and avoids skill swapping. It is an extremely potent feat (many would argue too potent). It single handedly boosts a lot of character concepts by existing, including propping up an entire sub class (Wit Swashbuckler). The parallel to Bon Mot is Demoralize, another CHA-based skill action that anyone has access to with different characteristics. Bon Mot essentially gives those who did not want to invest in Intimidation a suitable alternative, since skill increases are rather expensive and investing into Diplomacy and Intimidation both has some conceptual overlap (Coerce and Make an Impression fill similar narrative roles). Overall, Bon Mot isn’t perfect, it is arguably too strong and it doesn’t preclude high CHA or high skill increase classes from just investing into both Intimidation and Diplomacy (thus creating a higher character power overall), but it is bespoke, interesting, fun, and offers a concept that we have seen in pop culture for making witty comebacks mid fight to put your foe off balance.
Acrobatic Performer – This feat breaks the skill swap rule, but it does so under very clear precedents for doing so. For one, a player that is invested in Acrobatics likely is Dex focused and cannot afford to invest in Performance at all, which as a skill in general really only has applications for the Bard and to Earn Income. This essentially serves as a means for an Acrobat to use the Earn Income activity despite being uncharismatic without having to consistently invest in Performance just to present the concept of using your acrobatics to “Perform”. This does not change the power of the character, but allows them to engage in a narrative they otherwise should be able to despite their limitations.
Intimidating Prowess – This feat breaks the math fixer rule, but it does so because the characters that can qualify for it likely do not have an 18 in Charisma, and thus the ceiling on the ‘numbers’ for your Demoralize is the same as a Sorcerer despite being a fighter. This “math fixer” is essentially evening the playing field for martial combatants to engage in the Demoralize action economy, so it enables them to have the same level of power as what currently exists. It also contains a ribbon to ignore not sharing a language (since this is a visual presentation). Its subsequent upgrade at level 10 (master + 20) is once again to compete with CHA based casters as they also raise charisma.
Rolling Landing – This feat breaks the “prerequisite feat” rule, but it does so because the prerequisite feat is required for the resulting effect to even exist, since taking no damage from fall damage would be something exclusive to a person who uses Cat Fall anyways. This follow up feat builds on the concept by allowing extra movement on the fall (thus turning what is traditionally a “passive feat” in Cat Fall, to an “active feat” by allowing you to intentionally fall from out of view to move into position as part of that fall). It only has one skill feat as a prerequisite, distinctly builds on the outcomes of that feat instead of just “adding” to its power, and opens the feat from a passive position to an active one by having players able to plan around its usage instead of using it strictly to help prevent falling damage.
Natural Medicine – This feat breaks the skill swap rule, but it does so with reasonable outcomes. It replaces Medicine with another Wisdom based skill, so the swap “ceiling” for character concepts only serves to prevent over expenditure attempting to support both Nature and Medicine. Thematically, it also makes sense for natural remedies to offer similar medicine-like behaviors, so narratively the “swap” works conceptually.
Courtly Graces – This feat breaks the skill swap rule, but it does this in very specific circumstances that would make sense for a person with knowledge specifically about nobles. This allows INT based characters, who likely cannot afford CHA, to participate in a specific kind of communication and subterfuge that is not disruptive due to its scope of application. A character would still need Make an Impression for interacting with non-nobles and to disguise themselves as anyone but a noble. More importantly, it enables the concept of “I am not actually a noble, I’m just pretending to be one”, which is an extremely common trope in popular media. This collective design is what allows it to break the rule safely without encroaching on the niche of Charisma based characters, and allowing a specific character concept which otherwise for INT based characters is unavailable.
Quick Contacts – This feat breaks the rules for multiple prerequisites, however, it provides multiple avenues to achieve this route (Streetwise vs Courtly Graces, Connections vs Criminal Connections). Connections and Criminal Connections require an unspecified amount of time to build said connections, but presumably, you need to have been in that settlement for a good portion of time in order to use their effects. Quick Contacts applies a defined and finite downtime application (1 day) to both of those feats, thus making the previous applications more tangible and usable. All of the feats, including Quick Contacts, are level 2 or lower, which means that although they are “expensive” to achieve the end result of your feat chain of 3, they are all low enough level that a level 2 Rogue or any level 3 character could achieve all 3 of them with relative ease. I would argue that this, despite having reasons to break its rules, is not a good feat design (this is probably why it was not reprinted in Remaster and the previous mentioned feats were redefined under 1 umbrella of “Leverage Connections”. This seeks to give form to feats that already should have defined the amount of downtime it took to establish connections (Connections and Criminal Connections), but, it’s a campaign enabler which is probably what makes it so necessary. If your campaign takes place in the same city for most of the game, Quick Contacts is completely unnecessary, as you presumably would have connections in your current location without the need for quick swapping, but in other campaigns with large amounts of travel between settlements, it serves to give people access to those feats with more reliability and to unburden the GM from deciding when and how connections are made.
Disturbing Knowledge – This feat is a bespoke 2 action activity that does not break any of the rules, but sort of sidesteps the skill swap on a Demoralize via a unique action that’s similar. It is at the master level of Occultism, targets a single enemy within 30 feet, and applies immunity for 1 day regardless of the result (the legendary upgrade it applies to all within 30 feet). The effect itself is actually close to the same power level as Demoralize with Intimidating Glare, with the exception being that instead of frightened 2, you are confused for 1 round, and there is a critical failure effect on an opposed roll where you become frightened 1. Confused is certainly stronger than frightened 2 (arguably being confused and frightened 1 is not as bad as just being confused, since it applies to attacks that could target its allies) but the condition can end if they take damage and potentially frightening yourself is a big enough downside that it won’t disrupt play. However, this feat is merely “decent”. The 2-action cost makes it an expensive use (so it won’t be usable with 2-action spells), the master in Occultism means the person learning to do this activity likely has spells they could perform instead that would produce greater effects (albeit at the cost of resources), so ultimately this serves as a “resource conservation” feat and a way for Intelligence-based characters to participate in a thematic debuffer that they may not have access to otherwise. It is also pretty thematic and a common trope for “unspeakable knowledge” to cause such things, especially in existential horror books. When it gets its upgrade to Legendary, it becomes a very potent activity, but at that point if you have legendary Occultism and you are a caster, you have 8th-rank spells, and 3rd-rank Fear is better than this action every single time (and pretty cheap for a caster to use or have scrolls of at this point). For a non-caster specialized in Occultism, they get the most benefit out of this, but they also can’t use Occultism to Learn a Spell or for any of its other spell-related effects (such as activating magic items), and thus them benefiting from this in a higher capacity makes sense given their ability to use Occultism isn’t as potent as a caster.
All this to say that keeping things within the bounds of a skill feat and what the game expects skill feats to do can be a tricky game of ensuring you make something meaningful, while not encroaching too much on what Class Feats or even Ancestry Feats should be doing on their own. Using existing skill feats as a benchmark can often help you determine whether you are in bounds or out of bounds, especially when you want to break the “normal” rules.