Targeting¶

GAS includes a full targeting subsystem for abilities that need player-driven or automated target selection. It handles everything from "click to select a target" to "aim at the ground and show a radius indicator." The system is built from three main pieces:

Target Data — the payload that describes what was targeted (actors, locations, hit results)
Target Actors — the world actors that perform the actual targeting logic (traces, overlap checks, visual feedback)
Reticles and Filters — visual feedback during targeting and rules for which actors are valid targets

When to Use the Targeting System¶

The built-in targeting system is most useful when you need interactive targeting — the player aims, selects, confirms, and the ability reacts. Think MOBA-style ground targeting, RTS unit selection, or Diablo-style "point and click."

For simpler cases, you often do not need the full targeting system at all:

Scenario	What To Use
Melee attack hitting what is in front of the character	Line/sphere trace in the ability, wrap results in `FGameplayAbilityTargetData_SingleTargetHit`
Projectile that applies effects on hit	Trace or overlap in the projectile class, send results as a gameplay event
AoE around the caster	`FGameplayAbilityTargetData_ActorArray` with actors from an overlap check
Interactive ground targeting with preview	Full targeting system: Target Actor + Reticle + WaitTargetData task
Player selects a target with a crosshair	`AGameplayAbilityTargetActor_SingleLineTrace` with a reticle

Keep it simple

The targeting system is powerful but heavy — it spawns actors, manages replication, and has a multi-step confirm/cancel flow. If a simple trace in your ActivateAbility does the job, use that. Reach for the full system when you genuinely need interactive targeting with visual feedback.

The Three Pieces¶

Target Data — The Payload¶

FGameplayAbilityTargetData is a polymorphic struct that carries targeting results. It can hold hit results, actor lists, locations, or custom data. Target data is what flows from the targeting step to the effect application step.

Target Actors — The Targeting Logic¶

AGameplayAbilityTargetActor is a world actor spawned during targeting. It performs traces, overlap checks, or other targeting logic, and produces FGameplayAbilityTargetData when the player confirms. The built-in types cover line traces, ground traces, radius selection, and actor placement.

Reticles and Filters — Visuals and Validation¶

AGameplayAbilityWorldReticle provides visual feedback during targeting (the circle on the ground, the crosshair on the target). FGameplayTargetDataFilter controls which actors pass the targeting filter (exclude self, require a specific class, etc.).

The Targeting Flow¶

Here is how the pieces fit together in a typical interactive targeting scenario:

1. Ability activates
2. Ability spawns a WaitTargetData task
     └─ Task spawns or references a Target Actor
3. Target Actor runs targeting logic each tick
     └─ Shows a Reticle for visual feedback
     └─ Applies a Filter to candidate targets
4. Player confirms (or cancels)
5. Target Actor produces FGameplayAbilityTargetData
6. WaitTargetData task fires ValidData (or Cancelled) delegate
7. Ability receives target data and applies effects
8. Ability ends

The WaitTargetData ability task is the bridge between the ability and the target actor. It handles spawning the target actor, binding to its confirm/cancel delegates, and forwarding the resulting data back to the ability.

Targeting and Networking¶

Target data is designed to replicate. FGameplayAbilityTargetDataHandle implements NetSerialize and can be sent from client to server. The ShouldProduceTargetDataOnServer flag on target actors controls whether the server generates its own targeting data or trusts what the client sends.

For multiplayer games, you will typically:

Let the client run the targeting visualization
Send the confirmed target data to the server
Have the server validate and apply effects

The OnReplicatedTargetDataReceived virtual on target actors gives the server a chance to validate/sanitize incoming target data.