Use the verified spell boundary
More than thirty spells is an official scope statement, not permission to invent a list. The checked current sources do not establish every name, effect, cost, duration, cooldown, upgrade, or interaction. Those details are especially vulnerable to Early Access changes. Read the live in-game description for a choice and use this page to structure the decision around it. A framework can be useful without pretending to be a database.
If a later official post provides specific examples, this guide can cite and date them. Until then, a video clip, community post, or demo memory should be labeled by origin and build. Do not turn a visible result into an exact formula that was never measured or published. The claim boundary protects both beginners and returning players from advice that looks precise but has no current foundation.
Work from the interactions the official store names
The official description gives four concrete anchors: oil can be ignited, water can be electrified, a GHOUL can be polymorphed into a RAT, and Silent Step is a named stealth example. These details are useful for planning a squad conversation because they show that spells and class choices can interact with the bunker and with creatures. They do not establish a full combo catalog, exact costs, effect duration, or guaranteed result in every situation.
Treat each example as a prompt to inspect the current room and current in-game wording. Oil and water make the environment part of the decision. Polymorph points to control rather than a damage ranking. Silent Step gives stealth a verified example without proving a universal detection rule. Casting still consumes FAITH, so even a promising interaction belongs inside the same purpose, fallback, and reassessment loop as any other spell use.
Name the purpose before casting
A spell decision becomes easier to review when it has a declared purpose. Information spells would reduce uncertainty; movement choices would change position; control would shape a dangerous space; support would help another plan succeed; recovery would restore order after a mistake. These are editorial categories, not official spell schools. Their value is that the group can discuss an intended outcome without relying on unsupported names or numbers.
Ask whether the problem is immediate, whether positioning or an environmental interaction could answer it, and what happens if the spell does less than expected. The goal is not to avoid casting. It is to connect a FAITH spend to a reason and a fallback. In co-op, state the purpose so teammates can exploit the opening instead of duplicating the same response.
- Information: reduce uncertainty before commitment.
- Movement: create or preserve a route.
- Control: change how a space can be used.
- Support: strengthen another player’s action.
- Recovery: rebuild a workable state after disruption.
Use a flexible FAITH budget
Because casting consumes FAITH and lower FAITH exposes FEAR, the resource decision spans the whole run. Divide thinking into immediate need, expected near-term pressure, and protected reserve. These are not numerical breakpoints. They are questions. If a cast solves a minor inconvenience but removes the group’s answer to separation or danger, its opportunity cost is high. If delay would collapse the run, spending may preserve more options than saving.
Review the budget after major interactions, changes in group state, or new information about the bunker. A reserve chosen at the beginning should not become superstition. Equally, a successful aggressive spend does not prove that every future situation deserves it. Use the FAITH and FEAR guide to understand the system relationship, then let current observations revise the tactical budget.
Sequence squad spell decisions
Co-op adds coordination value and coordination risk. Two useful effects can be wasted if both answer the same moment without need, while a planned sequence can create information, movement, and recovery in order. Before a consequential cast, use a compact call: purpose, location, timing, and expected teammate action. The call does not need the whole mechanical explanation; it needs enough context for the group to avoid interference.
Designate no permanent spell commander unless the group wants one. The player with the clearest view can lead a local decision. Others should confirm whether they are following, holding, or proposing an alternative. Solo players can create the same sequencing discipline by pausing between effects and reassessing the environment instead of chaining actions based on the original room state.
Evaluate outcomes without fake precision
After a cast, compare the intended purpose with the observed result. Did it create a usable lane, reveal enough information, support the next action, or restore control? Note what else changed, including position and environmental state. Avoid declaring exact damage, duration, or universal efficiency when the source does not provide it. A qualitative observation with version context is more honest than a number copied from an unknown build.
One outcome is not a ranking. Procedural layouts and team execution can make the same option feel different. Collect repeated observations and distinguish them from official facts. This site does not claim first-hand testing, so its recommendations remain decision frameworks grounded in official system descriptions. Readers should use the current build to verify the details visible to them.
Keep spell notes build-aware
When Early Access updates arrive, recheck in-game wording before relying on saved notes. Record the date or version alongside any discovered interaction. If the wording and behavior differ, treat that as something to verify, not immediate proof of a present defect. The known-issues guide explains how official notices, current community reports, and historical demo context must be separated.
A durable spell note contains purpose, situation, observed outcome, and build context. It avoids unsupported superlatives. Share that context with the group before a run so returning players know which assumptions may be stale. The builds page can then combine spell purposes with class responsibilities without pretending that one composition is universally correct.