mcpbeat

Unreal Blueprints

gamedev-skills/unreal-blueprints

> Event Graph and Construction Script, variables/functions/macros, and Blueprint communication (Cast, Interfaces, Event Dispatchers). Use when working in Blueprints, wiring an event graph, deciding how Blueprints talk to each other, or when the user mentions Blueprint, BP, event graph, construction script, or a Blueprint .uasset.

2k tokens
context cost
the whole folder, loaded on every use
2
files
instructions only
0
copies elsewhere
how many repositories repackaged it
401
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/gamedev-skills/awesome-gamedev-agent-skills --skill unreal-blueprints

The instruction itself

11 sections, as written by the author

Unreal Blueprints (Visual Scripting)

Structure gameplay logic in Unreal Engine 5 Blueprints: pick the right graph, expose data

cleanly, and choose a communication method that doesn't create hard-reference spaghetti.

Targets UE 5.4+. (Blueprints are node graphs; the snippets below describe node flows.)

When to use

  • Use when authoring a Blueprint Class, wiring the Event Graph (BeginPlay/Tick/overlap),

using the Construction Script, creating variables/functions/macros, or choosing how two

Blueprints communicate (Cast, Interface, or Event Dispatcher).

  • Use when the project has *.uproject and Blueprint *.uasset files, and the user works

visually rather than in C++.

When *not* to use: performance-critical systems, large data structures, or anything that

benefits from source control diffs and unit tests → unreal-cpp-gameplay. Player input

mapping → unreal-enhanced-input. AI logic → unreal-behavior-trees.

Core workflow

  • Choose the Blueprint type. A Blueprint Class (derived from Actor/Pawn/Character/

ActorComponent) defines a reusable object. The Level Blueprint is a per-level graph for

level-specific scripting only — don't put reusable logic there.

  • Use the Construction Script for editor-time setup (procedural placement, configuring

components from variables) — it runs when the actor is placed or edited, *not* during play.

  • Use the Event Graph for runtime logic. Event BeginPlay for init, input/overlap events

for reactions. Avoid Event Tick unless you truly need per-frame work.

  • Expose data with variables; click the eye icon to make a variable Instance Editable, and

group related ones with categories. Mark pure functions (no exec pin) for getters.

  • Pick a communication method by coupling (see Patterns): direct Cast for things you

own, Blueprint Interface to call across types without hard references, **Event

Dispatcher** to broadcast one-to-many.

  • Verify with the Blueprint debugger: drop breakpoints on nodes, watch variable values,

and use Print String to confirm execution paths during Play In Editor (PIE).

Patterns

1. Reactive Event Graph (no Tick)

Event BeginPlay
  -> Set 'StartLocation' = GetActorLocation
  -> Bind Event to OnComponentBeginOverlap (TriggerVolume) [calls custom event OnEnterZone]

OnEnterZone (Other Actor)
  -> Branch: Other Actor == Player?
       True  -> Open Door (Timeline drives the rotation)   // event-driven, runs once

Prefer events (overlaps, timers, dispatchers) and Timelines over polling in Tick.

2. Direct reference + Cast (tight coupling, use sparingly)

Overlapped Actor (Actor ref)
  -> Cast To BP_Player
       Cast Failed -> (do nothing)
       Success     -> call BP_Player.ApplyDamage(10)

Cast To creates a hard reference to that class (it loads with this Blueprint). Fine when the

caller genuinely depends on that type; otherwise prefer an Interface.

3. Blueprint Interface (decoupled call)

// 1. Create BPI_Interactable with function 'Interact(Instigator)'.
// 2. Add the interface to BP_Door, BP_Chest, BP_Lever and implement 'Interact' in each.
// 3. Caller, with any Actor ref:
Player presses Use
  -> Does Object Implement Interface (BPI_Interactable)?  // safe check, no Cast/hard ref
       True -> Interact (Message) on Target Actor

4. Event Dispatcher (one-to-many broadcast)

// In BP_Player: declare Event Dispatcher 'OnHealthChanged (float NewHealth)'.
TakeDamage -> Set Health -> Call 'OnHealthChanged' (Health)   // broadcast

// In WBP_HUD BeginPlay: Bind Event to 'OnHealthChanged' -> update health bar.
// Many listeners can bind; the player never references them.

Pitfalls

  • Cast spaghetti / long load times — chains of Cast To create hard references that pull

whole asset trees into memory. Decouple with Interfaces or Dispatchers.

  • Logic in the Level Blueprint that should be reusable — it can't be reused across levels.

Put it in a Blueprint Class.

  • Event Tick overuse — every-frame nodes add up fast. Use events, Timers

(Set Timer by Event), and Timelines instead.

  • Construction Script doing gameplay — it runs in the editor on edit/placement; spawning

gameplay actors or starting logic there causes editor-only artifacts. Init in BeginPlay.

  • Variable not visible on the instance — toggle Instance Editable (the eye); to edit before

spawn via Spawn node, also mark "Expose on Spawn".

  • Interface call did nothing — the target doesn't implement the interface; use "Does

Implement Interface" before calling, or use the Message version which is safe on non-implementers.

References

  • For a decision guide on Cast vs Interface vs Event Dispatcher and step-by-step dispatcher

binding, read references/communication.md.

  • Primary docs: "Blueprints Visual Scripting"

(https://dev.epicgames.com/documentation/en-us/unreal-engine/overview-of-blueprints-visual-scripting-in-unreal-engine).

  • unreal-cpp-gameplay — when to drop to C++; how BP and C++ classes interoperate.
  • unreal-enhanced-input — the modern way to feed input events into these graphs.
  • unreal-behavior-trees — AI decision logic that Blueprints trigger.

How to use it

Copy the folder

Take gamedev-skills/unreal-blueprints from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.