Customizable Object logo

Customizable Object

Community
kevinpbuckley
customizable-object

Read, build, wire, parameterize and compile Mutable Customizable Object graphs (UCustomizableObject) — skeletal mesh chains, mesh sections and materials, components, enum/float/color/texture parameters, switches, groups and child objects, compile errors (CustomizableObjectService). Use when the user asks about Mutable, a Customizable Object / CO asset, character customization graphs, or modular characters built with Mutable.

Overview

Publisherkevinpbuckley
RepositoryVibeUE
Skill namecustomizable-object
Stars
735
Forks
161
Bundled files
1
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by kevinpbuckley on GitHub. Read the source before you install it.

Installation

Install the Customizable Object AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/kevinpbuckley/VibeUE/tree/master/Content/Skills/customizable-object
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/kevinpbuckley/VibeUE.git /tmp/VibeUE
mkdir -p .claude/skills
cp -r /tmp/VibeUE/Content/Skills/customizable-object .claude/skills/customizable-object
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Customizable Object in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Customizable Object on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Customizable Object is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

🧠 Brains complement: IF an unreal-engine-skills-manager tool (external MCP) exists in this session, call it with {action: "search", query: "mutable customizable object"} for UE domain knowledge on Mutable. If no such tool is available, skip this line entirely and proceed with this skill alone — do NOT attempt the call.

Mutable Customizable Objects Skill

One service, reachable from Python (unreal.CustomizableObjectService.*) and as AICallable tools. It edits the Customizable Object's editor graph (the node graph you see in the CO editor) and compiles it. Discover exact signatures with discover_python_class('unreal.CustomizableObjectService').

  • Is Mutable on? is_mutable_available(). In UE 5.8 it normally is (the default-enabled MutableAssetUserData plugin loads it). If it returns False, every method explains that the project has Mutable disabled — tell the user; VibeUE never enables plugins for them.
  • Addressing: nodes by node_id (the NodeGuid, from get_graph / add_node); pins by pin_name OR the display name the editor shows ("Mesh" works for Mesh_Input_Pin).
  • Every write saves the asset and returns success / error. Read error — it names the cause (incompatible pin categories, unknown property, value that did not parse).

The golden workflow — a compilable object

python
import unreal, json
CO = unreal.CustomizableObjectService
path = "/Game/Characters/CO_Hero"
MESH = "/Game/Characters/Meshes/SK_Body.SK_Body"          # any skeletal mesh WITH a skeleton
MAT  = "/Game/Characters/Materials/M_Body.M_Body"

base = CO.create_customizable_object(path)                # graph + Base Object node
b = base.node_id

# Nodes. Pass properties that shape pins AT CREATION (see "Pins depend on properties" below).
sk   = CO.add_node(path, "CustomizableObjectNodeSkeletalMesh", -1200, 0, json.dumps({"SkeletalMesh": MESH}))
sec  = CO.add_node(path, "CONodeSkeletalMeshSection",      -900, 0, json.dumps({"Material": MAT}))
make = CO.add_node(path, "CONodeSkeletalMeshMake_V2",      -600, 0)
obj  = CO.add_node(path, "CONodeSkeletalMeshObjectMake",   -400, 0)
comp = CO.add_node(path, "CONodeComponentSkeletalMesh",    -200, 0,
                   json.dumps({"ReferenceSkeletalMesh": MESH, "ComponentName": "Body"}))

# Wiring: output -> input (either order is accepted).
CO.connect_pins(path, sk.node_id,   "LOD 0 - Section 0",         sec.node_id,  "Mesh")
CO.connect_pins(path, sec.node_id,  "Mesh Section",              make.node_id, "LOD 0")
CO.connect_pins(path, make.node_id, "Skeletal Mesh",             obj.node_id,  "Skeletal Mesh")
CO.connect_pins(path, obj.node_id,  "Passthrough Skeletal Mesh", comp.node_id, "Passthrough Skeletal Mesh")
CO.connect_pins(path, comp.node_id, "Component",                 b,            "Components")

r = CO.compile_object(path)
print(r.success, r.state, list(r.errors), list(r.warnings))
info = CO.get_object_info(path)                           # compiled parameters, states, components

That chain — Skeletal Mesh → Skeletal Mesh Section → Skeletal Mesh Make V2 → Skeletal Mesh Object Make → Component Skeletal Mesh → Base Object "Components" — is the minimal compilable object in UE 5.8. Everything else (parameters, switches, modifiers, children) hangs off it.

Reading an existing object

python
g = CO.get_graph(path)                 # nodes (class, title, pins with links) + flat connections
for n in g.nodes: print(n.node_id, n.node_class, n.title)
CO.get_node_properties(path, node_id)  # settable properties with current values (text format)
CO.get_object_info(path)               # parent/children, graph + compiled parameters, states, components
CO.discover_node_types("Parameter")    # node classes add_node accepts (filter is a substring)
CO.get_node_type_properties("CONodeSwitch")   # a class's properties and defaults

Things that are not obvious and will bite you

  • A Section needs a material. Without Material on the CONodeSkeletalMeshSection the object still compiles, but with the warning "Could not generate a mesh section because it didn't have a material selected" — and that section is missing from the result. Set it at creation or with set_node_property(path, sec, "Material", "/Game/.../M_X.M_X"), or link a Material node into its Material pin.
  • The component needs ReferenceSkeletalMesh, and that mesh needs a Skeleton. Without it the compile FAILS: "Missing reference Skeletal Mesh".
  • Pins depend on properties. A Skeletal Mesh node has one output per LOD/section ("LOD 0 - Section 0 - Mesh", display "LOD 0 - Section 0") only once SkeletalMesh is set; a Switch's output type comes from PinType. Pass these in add_node's properties_json, or set them afterwards with set_node_property (the node rebuilds its pins and the result lists them). PinType on a switch must be given at creation.
  • Switches: create CONodeSwitch with {"PinType": "<category>"} — material (mesh section), mesh, color, float, image, component, … — then link an Enum Parameter's Enum output into its Switch Parameter input. It grows one input per enum value, named after the value. Editing the enum's Values later rebuilds the switch automatically.
  • An unconnected switch option compiles silently. Add an enum value and the switch grows an input for it; leave that input empty and the compile reports no error and no warning, yet the option has nothing behind it. After editing enum values, check every switch input is linked (get_node(path, switch).pins[*].linked_to).
  • Compile problems are not on the nodes. get_graph nodes' message stays empty; read compile_object(...).errors / .warnings, which end with (Node <title>).
  • Enum values are a struct array: set_node_property(path, enum, "Values", '((Name="Red"),(Name="Blue"))'). Read get_node_properties first and mirror the shape it shows for any struct/array property.
  • Parameter names are properties, not pins: ParameterName (parameters), ObjectName (Object nodes), GroupName (groups), ComponentName (components). They are listed with editable=False because the editor uses custom UI for them — they are still settable.
  • Compiled vs graph parameters. get_object_info().graph_parameters lists parameter nodes in this asset right away; compiled_parameters only reflects the last successful compile and only includes parameters that feed the object. Recompile after edits before trusting it. An enum parameter compiles to type Int.
  • Numbers and booleans are validated. "abc" for an int is refused (the engine would store 0).
  • Pin types must match. connect_pins refuses incompatible categories with the schema's reason; a single-link input silently dropping its old link is reported in notes.
  • The Base Object node cannot be deleted. Tunnel nodes (Macro Library only) cannot be added.

Groups and child objects

A Group node (CustomizableObjectNodeObjectGroup) is a parameter over child objects: GroupType = COGT_TOGGLE / COGT_ALL / COGT_ONE / COGT_ONE_OR_NONE, named by GroupName. Link its Group output to the Base Object's Children input. Children live in their own assets:

python
grp = CO.add_node(root, "CustomizableObjectNodeObjectGroup", 0, 300,
                  json.dumps({"GroupName": "Hat", "GroupType": "COGT_ONE_OR_NONE"}))
CO.connect_pins(root, grp.node_id, "Group", root_base_id, "Children")

cap = CO.create_customizable_object("/Game/Characters/CO_Hat_Cap")
CO.set_node_property("/Game/Characters/CO_Hat_Cap", cap.node_id, "ObjectName", "Cap")
CO.set_parent_object("/Game/Characters/CO_Hat_Cap", root, grp.node_id)   # "" parent detaches
# ...build the cap's own mesh chain into ITS base node's Components...
CO.compile_object(root)     # compile the ROOT; the "Hat" parameter gains the option "Cap"

The child's ObjectName becomes its option name. get_object_info(root).child_objects lists the children; parenting refuses cycles and non-Group targets.

Compile

compile_object(path, optimization_level="None", texture_compression="Fast") is synchronous and always a real compile. errors / warnings are the compiler's own messages (most end with (Node <title>) naming the culprit); log has every Mutable log line. success needs a valid model AND no errors. Compiling a child compiles its root.

Deeper reference

customizable-object/node-reference — the common node classes with their pins and properties.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Customizable Object AI skill do?

Read, build, wire, parameterize and compile Mutable Customizable Object graphs (UCustomizableObject) — skeletal mesh chains, mesh sections and materials, components, enum/float/color/texture parameters, switches, groups and child objects, compile errors (CustomizableObjectService). Use when the user asks about Mutable, a Customizable Object / CO asset, character customization graphs, or modular characters built with Mutable.

Why use Customizable Object on TypingMind?

Because you install it once and use it with any model. Customizable Object is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Customizable Object in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/kevinpbuckley/VibeUE/tree/master/Content/Skills/customizable-object. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Customizable Object?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Customizable Object?

As many as you like. As long as a model supports skills, you can use Customizable Object with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Customizable Object AI skill free?

Yes. It is published on GitHub by kevinpbuckley under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇