Sound Cues logo

Sound Cues

Community
kevinpbuckley
sound-cues

Create and modify SoundCue assets — add/connect nodes (mixer, random, delay, attenuation, modulator, etc.) and set audio properties (SoundCueService). Use when the user asks to create a Sound Cue, build a SoundCue node graph, or wire audio playback logic.

Overview

Publisherkevinpbuckley
RepositoryVibeUE
Skill namesound-cues
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 Sound Cues 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/sound-cues
  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/sound-cues .claude/skills/sound-cues
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Sound Cues 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 Sound Cues 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 Sound Cues 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: "load", skill: "audio-and-metasounds"} for UE domain knowledge on this topic — correct APIs, architecture, best practices — and treat it as the rubric for any review / "best practices" question. If no such tool is available (e.g. running under Claude Code or Codex without that MCP), skip this line entirely and proceed with this skill alone — do NOT attempt the call.

Wrong skill for MetaSound assets? If the user asked about MetaSound, MS_ assets, or UMetaSoundSource, load the metasounds skill instead (via GetSkills / AgentSkillToolset). SoundCue and MetaSound are completely separate systems — do not use SoundCueService for MetaSound tasks.

Sound Cue Editor Skill

Service Access

python
import unreal
svc = unreal.SoundCueService()

Audio Flow Direction

Audio flows FROM leaves TOWARD root.

[WavePlayer] ──> [Random] ──> [Mixer] ──> [ROOT]
                                ^
                    [WavePlayer2] ──────┘
  • connect_nodes(parent, child, slot) means: child provides audio INTO parent at that input slot.
  • set_root_node(index) makes that node the final output (what you hear).

Critical Rules

⚠️ Bool Properties — b prefix stripped in Python

UE strips the b prefix from bool UPROPERTY names in Python:

C++ / UPROPERTYPython attribute
bSuccesssuccess
bIsRootNodeis_root_node
bLoopingis_looping
bStreamingis_streaming

⚠️ Node indices are positional and may shift

list_nodes() returns nodes in graph order — indices can change if new nodes are added. Always call list_nodes() to get current indices before connecting or modifying nodes.

⚠️ Quality Level node has a fixed slot count

add_quality_level_node creates a node whose input-slot count is governed by the project's audio quality levels (AudioQualitySettings), not by an argument — there is no num_inputs param. In a default/skeleton project this is often a single slot, so connect_nodes(..., slot=1) returns False. If a cue must select between sounds per scalability bucket, confirm the project defines multiple quality levels first; otherwise only slot 0 is wireable.

⚠️ New-cue defaults: volume_multiplier is 0.75, not 1.0

A freshly created empty cue reports volume_multiplier = 0.75 (the engine USoundCue default), pitch_multiplier = 1.0. Do not assert 1.0 for volume on a new cue — read it back instead.

⚠️ connect_nodes parameter order

python
# parent receives audio FROM child at slot
svc.connect_nodes(cue_path, parent_index, child_index, input_slot)

# CORRECT: WavePlayer(0) feeds into Random(1) at slot 0
svc.connect_nodes('/Game/MyCue', 1, 0, 0)

# WRONG: don't confuse child/parent order

Workflows

Create a Simple SoundCue

python
import unreal
svc = unreal.SoundCueService()

# Create the asset (optionally wire an initial WavePlayer)
r = svc.create_sound_cue('/Game/Audio/SC_Footstep', '')
assert r.success, r.message

# Add a WavePlayer node
r = svc.add_wave_player_node('/Game/Audio/SC_Footstep', '/Game/Audio/Footstep_01', -250, 0)
assert r.success

# Set it as the root (output)
nodes = svc.list_nodes('/Game/Audio/SC_Footstep')
svc.set_root_node('/Game/Audio/SC_Footstep', nodes[0].node_index)

# Save
svc.save_sound_cue('/Game/Audio/SC_Footstep')

Random from Multiple Waves

python
import unreal
svc = unreal.SoundCueService()
cue = '/Game/Audio/SC_Footsteps'

svc.create_sound_cue(cue, '')

# Add 3 wave players
for i, wave in enumerate(['/Game/Audio/Step_01', '/Game/Audio/Step_02', '/Game/Audio/Step_03']):
    svc.add_wave_player_node(cue, wave, -400, i * 150)

# Add a Random node to select one
svc.add_random_node(cue, -150, 150)

nodes = svc.list_nodes(cue)
# nodes[0..2] = WavePlayers, nodes[3] = Random
random_idx = next(i for i, n in enumerate(nodes) if 'Random' in n.node_class)
for i in range(3):
    svc.connect_nodes(cue, random_idx, i, i)  # each wave into random

svc.set_root_node(cue, random_idx)
svc.save_sound_cue(cue)

Mixer (parallel blend)

python
import unreal
svc = unreal.SoundCueService()
cue = '/Game/Audio/SC_Ambient'

svc.create_sound_cue(cue, '')
svc.add_wave_player_node(cue, '/Game/Audio/Wind', -400, 0)
svc.add_wave_player_node(cue, '/Game/Audio/Rain', -400, 150)
svc.add_mixer_node(cue, 2, -150, 75)   # 2 inputs

nodes = svc.list_nodes(cue)
wave0 = next(i for i, n in enumerate(nodes) if 'WavePlayer' in n.node_class and i == 0)
wave1 = next(i for i, n in enumerate(nodes) if 'WavePlayer' in n.node_class and i != wave0)
mixer = next(i for i, n in enumerate(nodes) if 'Mixer' in n.node_class)

svc.connect_nodes(cue, mixer, wave0, 0)
svc.connect_nodes(cue, mixer, wave1, 1)
svc.set_root_node(cue, mixer)
svc.save_sound_cue(cue)

Volume / Pitch / SoundClass

python
svc.set_volume_multiplier('/Game/Audio/SC_Explosion', 1.5)
svc.set_pitch_multiplier('/Game/Audio/SC_Explosion', 0.9)
svc.set_sound_class('/Game/Audio/SC_Explosion', '/Game/Audio/SC_SFX')

API Reference

Asset Lifecycle

MethodReturnsNotes
create_sound_cue(path, wave_path)FSoundCueResultwave_path='' for empty cue
duplicate_sound_cue(src_path, dst_path)FSoundCueResultcopy a cue to a new path
delete_sound_cue(path)booldeletes the asset from disk
get_sound_cue_info(path)FSoundCueInfonode count, root index, vol, pitch, duration
save_sound_cue(path)boolsaves to disk

Node Creation — all return FSoundCueResult

MethodNode TypeSpecial params
add_wave_player_node(path, wave_path, x, y)Leaf — plays a SoundWavewave_path optional
add_random_node(path, x, y)Picks one child randomly—
add_mixer_node(path, num_inputs, x, y)Blends children in parallelnum_inputs 1–32
add_concatenator_node(path, num_inputs, x, y)Plays children in sequencenum_inputs 2–32
add_modulator_node(path, x, y)Random pitch/volume variance—
add_attenuation_node(path, x, y)Spatial attenuation (graph node)—
add_looping_node(path, x, y)Loops its child—
add_delay_node(path, x, y)Adds a delay before playing—
add_switch_node(path, x, y)Routes by integer parameter—
add_enveloper_node(path, x, y)Volume/pitch envelope over time—
add_distance_cross_fade_node(path, num_inputs, x, y)Crossfades by distancenum_inputs 2–32
add_branch_node(path, x, y, bool_param='')Routes by bool parameter; True(0) False(1) Unset(2)optional param name last
add_param_cross_fade_node(path, num_inputs, x, y)Crossfades by named float paramnum_inputs 2–32
add_quality_level_node(path, x, y)One input per quality levelslot count = project quality levels (see note)

Node Management

MethodReturnsNotes
list_nodes(path)TArray<FSoundCueNodeInfo>index, class, children, is_root_node
connect_nodes(path, parent, child, slot)boolchild feeds audio INTO parent
disconnect_node(path, parent_index, slot)boolbreak a single input link
set_root_node(path, index)boolsets cue output node
set_wave_player_asset(path, index, wave_path)boolreassign wave on existing node
remove_node(path, index)booldelete a node from the graph
move_node(path, index, x, y)boolreposition a node in the graph
get_node_property(path, index, prop_name)FStringread any node UPROPERTY
set_node_property(path, index, prop_name, value)boolwrite any node UPROPERTY

Cue-Level Settings

MethodReturnsNotes
set_volume_multiplier(path, vol)bool—
set_pitch_multiplier(path, pitch)bool—
set_sound_class(path, class_path)bool—
set_attenuation(path, att_path)boolatt_path='' to clear
get_attenuation(path)FStringreturns asset path or ''
set_concurrency(path, conc_path, clear=False)boolconc_path='' + clear=True to clear all
get_concurrency(path)TArray<FString>list of assigned concurrency asset paths

FSoundCueNodeInfo fields

python
n.node_index      # int — position in list_nodes array
n.node_class      # str — e.g. "SoundNodeWavePlayer"
n.node_title      # str — display name
n.pos_x, n.pos_y  # float — graph position
n.is_root_node    # bool — True if this node is SoundCue.FirstNode
n.child_indices   # list[int] — indices of connected children (-1 = unconnected slot)

SoundWave Utilities

MethodReturnsNotes
get_sound_wave_info(sw_path)FSoundWaveInfoduration, sample rate, channels, looping, streaming
import_sound_wave(file_path, asset_path)FSoundCueResultimport .wav/.mp3 from disk
set_sound_wave_property(sw_path, prop, value)boolset any UPROPERTY by name

Common set_sound_wave_property property names:

PropertyNameTypeExample value
bLoopingbool"true"
bStreamingbool"true"
VolumeMultiplierfloat"1.5"
PitchMultiplierfloat"0.9"
SubtitlePriorityfloat"100.0"

FSoundWaveInfo fields

python
info.asset_path    # str
info.duration      # float — seconds
info.sample_rate   # int — Hz
info.num_channels  # int — 1=mono, 2=stereo
info.looping       # bool  (note: NOT is_looping — UE strips the 'b' only)
info.streaming     # bool  (note: NOT is_streaming)

FSoundCueResult fields

python
r.success      # bool
r.asset_path   # str
r.message      # str — human-readable status or error

Sample scripts (run via execute_python_code)

  • scripts/create_sound_cue.txt — create a SoundCue (optionally from a wave) and add nodes.

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 Sound Cues AI skill do?

Create and modify SoundCue assets — add/connect nodes (mixer, random, delay, attenuation, modulator, etc.) and set audio properties (SoundCueService). Use when the user asks to create a Sound Cue, build a SoundCue node graph, or wire audio playback logic.

Why use Sound Cues on TypingMind?

Because you install it once and use it with any model. Sound Cues 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 Sound Cues in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/kevinpbuckley/VibeUE/tree/master/Content/Skills/sound-cues. 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 Sound Cues?

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 Sound Cues?

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

Is the Sound Cues 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 👇