Aura can plan, create, and edit Blueprints. This page outlines the different ways Aura can help with Blueprints.
Aura can help you plan a Blueprint from start to finish. We recommend planning with Aura in Ask mode first before having Aura make edits. Planning first allows for quicker iteration with Aura and can save rework later on. Once you think a plan looks good, you can easily switch to Agent mode and have Aura follow the plan (see Blueprint Agent).
Examples

Aura can run comprehensive code reviews on your blueprints, like having a expert engineer on your team. It’ll analyze for typos, hanging nodes, logic errors and more. Code reviews focus on static analysis, logic errors, and convention. It is not meant to debug issues with your gameplay.
You can run reviews in two ways:


Reviews can take several minutes currently. Remember you can work in multiple chat threads at the same time!
Remember that Aura can make mistakes, always check the answer.
💡 Have particular naming conventions or styles for your blueprints?
Try setting up Custom Instructions that Aura will follow every time. Learn more in Advanced Settings
The crash course below utilizes Blueprint planning and generation.
Aura can now create and edit blueprints for you. Blueprint generation is still in early development and so Aura can’t one-shot complex systems (see Support Systems), but it can certainly help you get started.

⚠️ By default Edit Existing Blueprints is turned on. You can toggle this in Settings > Chat Options > Toggle on “Edit Existing Blueprints”

What we’ve found works well for this alpha version:
This workflow is faster than merging a huge graph and trying to debug everything afterward.
Avoid trying to one-shot a system. Aura is more effective when you break work into smaller parts that can be easily tested along the way.
Instead of asking Aura:
“Create an inventory system.”
Break it into small components like:
Data definitions:
“Help me define the item data I need (ID, name, icon, max stack, type) and advise whether it should live in Data Assets, Data Tables, or structs.”
Inventory logic:
“Outline the core functions for an Inventory Component (Add/Remove/Has/GetQuantity) and edge cases I should handle (overflow, full inventory).”
UI layer:
“Give me a high-level UI approach for a grid/list inventory that reads from the Inventory Component and refreshes via events.”
Interaction flow:
“Describe a pickup/drop flow (world pickup actor → interact → add → update/destroy pickup) for the player character.”
Persistence:
“What inventory data should go into a Save Game object, and how do I restore it on load?”
This approach helps you:
If you want a quick first win, start with a small system like a jump pad, interactable pickup, or basic spawner, then scale up from there.
Aura works best when your intent is clear, specific, and describes a trigger or condition.
Good vs. Bad Prompts
BP_Enemy, call ApplyDamage on it.”OnHit triggers, spawn VFX FX_Explosion and play sound SFX_Hit.”It’s also important to keep scope manageable. Aura performs best when working with 1-3 blueprints at a time, making it ideal for small game systems. Aura also currently performs better making new functions/events from scratch over editing existing complicated graphs.
💡 Blueprint Graph Agent is still in alpha. Be sure to save and backup your project regularly.
Aura can create and configure Timeline nodes on actor blueprints. Describe the animation or event sequence you need and Aura will build the Timeline, add the appropriate tracks, and wire it into your Blueprint graph.
Aura supports all four Timeline track types: float, vector, linear color, and event.
Examples
💡 Be specific about duration and whether the Timeline should autoplay or loop. "A 2-second looping float track that goes from 0 to 1" gives Aura a clear target to hit.
We’re actively working on improving Aura’s blueprint capabilities. It’ll only get better from here on out!
As of Aura v0.13.7, Aura can handle prompts like:
Aura is still evolving but even today it can create blueprints faster than any other tool out there. We’re actively working on improving its vocabulary of nodes and supported systems.