Audio Direction logo

Audio Direction

Organization
SummerEngine
audio-direction

Define the game's sound (music style, SFX vocabulary, dynamic music) as .summer/audio-bible.md.

Overview

PublisherSummerEngine
Repositorysummer
Skill nameaudio-direction
Stars
91
Forks
8
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 SummerEngine on GitHub. Read the source before you install it.

Installation

Install the Audio Direction 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/SummerEngine/summer/tree/main/library/skills/audio-direction
  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/SummerEngine/summer.git /tmp/summer
mkdir -p .claude/skills
cp -r /tmp/summer/library/skills/audio-direction .claude/skills/audio-direction
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Audio Direction 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 Audio Direction 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 Audio Direction 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.

/audio-direction — Define the Sonic Identity

Overview

Many indie games sound like one sound library used flat: the same generic "swoosh" for menu, jump, pickup and hit. This skill builds an audio bible that names the music style, the SFX vocabulary, the dynamic music plan, and the spatial / mix rules. Writes to .summer/audio-bible.md and configures the audio bus layout in the project so the rest of the work has a real foundation.

Core principle: SFX needs a vocabulary, not a library. Eight named SFX classes with consistent character beat 80 random library hits.

Steps

1. Open with three references

Ask exactly this:

Three reference tracks, games, or films for the music. What instruments matter? And what's the SFX vibe — chunky, clean, synthetic, or organic?

Three is the minimum. One or two collapses into a clone.

If they freeze, offer four axes (not titles):

Pick the closest: a) orchestral + cinematic (Hollow Knight, Journey), b) lo-fi electronic + ambient (Disco Elysium, Outer Wilds), c) chiptune / synth (Hyper Light Drifter, Celeste), d) acoustic / folk (Sable, A Short Hike).

2. Read the brief and art bible if they exist

Read .summer/GameSoul.md
Read .summer/art-bible.md

Audio must rhyme with the visuals. Quote it back:

Art bible says "soft, hand-warmed, golden-hour, painterly". That rhymes with acoustic / folk / lo-fi over orchestral or aggressive synth. Confirming.

If the brief says "horror", that constrains the music style harder than the user's references will. Name the constraint.

3. Pin the music style

Force one primary style. Sub-styles can blend, but a primary is required.

StyleAnchor instrumentsPairs withAvoid
Orchestral cinematicStrings, brass, choirStory-led, fantasyCasual / cozy
Acoustic / folkGuitar, piano, hand percussionCozy, narrativeAction / horror
Lo-fi electronicVintage synth pad, tape sat, light beatAtmospheric, indieCombat-heavy
Chiptune / synthwave8-bit / 16-bit chips, FM synthRetro, platformerRealism
Industrial / metalDistorted bass, percussion-ledHorror, actionCozy
Ambient dronePads, no drums, slow movementWalking sim, horrorAction
Jazz / fusionLive instruments, improvStylish, narrativeGeneric action
Tracker / hyperpopGlitchy, dense, modernEsoteric, indieMass appeal

Ask:

One primary style. Which?

4. Define tempo + key range

Concrete numbers, not "varies".

Tempo range: 80–110 BPM (slow side; combat would push 120+)
Key center: D minor (warm but melancholic)
Modulation rule: combat tracks shift to D dorian, calm tracks to D natural minor
Time signature: 4/4 default; 6/8 for emotional cues

5. Build the SFX vocabulary

Eight named classes. Each class has a character, not a single sound.

ClassCharacterExamples in this gameDON'T
UISoft, woody, short (60-100ms)Menu hover, button click, tab switchNo bell tones, no synth blips
Pickup positiveWarm, ascending interval (major third), 200-400msCoin, key, healNo pure sine "ding"
Pickup neutralBrief tactile, no melodic componentNote, breadcrumb, lore itemNo ascending pitch (reads as reward)
Negative / damagePunchy low-mid, slight detunePlayer hit, enemy hit, fall damageNo long reverb tail
Movement / footstepMaterial-aware (wood / stone / grass)Walk, run, jump, landNo same sample for all surfaces
Attack / actionFront-loaded transient, short bodySwing, parry, dashNo layered orchestral hit
Ambient layerContinuous, low presenceWind, room tone, distant wavesNo looping melody
Stinger / eventDistinctive, rare, melodicBoss intro, level clear, story beatNo reuse for normal events

Ask:

Eight classes above. Adjust any character / examples / DON'Ts?

6. Define the dynamic music plan

Static music = one track loops forever in a level. Boring. Dynamic music = layers swell with player state. Pick a model.

ModelHow it worksCostBest for
Layered stemsOne bar / multiple instrument layers; combat adds drum stem, calm removes itMedium (need stems authored)Most games
Vertical re-orchestrationSame melody, different instrumentation per stateHigh (composer-led)Story-led / boss-heavy
Horizontal sequencingDifferent tracks for different states, crossfadeLow (just author 3-5 full tracks)Vertical slice scope
Procedural / generativeGenerate from rules in real-timeHigh dev costSpecial-case

Pick one. State the state machine:

States:
  CALM        — exploration, low intensity
  TENSION     — enemies aware but not engaged
  COMBAT      — engaged
  VICTORY     — encounter cleared (4-second sting → return to CALM)

Transitions:
  CALM → TENSION   when an enemy detects the player (2s crossfade)
  TENSION → COMBAT when first damage exchanged (instant drum stem)
  COMBAT → VICTORY when last enemy dies (drum stem cuts, brass swell)
  VICTORY → CALM   after 4s

Rules:
  No state stays > 90s without variation (loop with bar-1 randomization)
  Volume ducking: dialogue active → music -6dB
  Boss music is its own track; ignores normal state machine

7. Define the spatial + mix rules

RuleValue
3D attenuation curveInverse-square (default), max distance 25m
Reverb buses1 small (interior), 1 large (cave / hall), 1 outdoor
HRTF for playerON (headphone-friendly directional cues)
Master ceiling-1 dBTP (true peak, no clipping on consoles)
Music vs SFX vs UI mix-8 dB / 0 dB / -3 dB (SFX is loudest)
Footstep rolloffCuts at 4m for player; cuts at 10m for NPC (player needs them, distant NPCs are noise)
CompressionLight bus compression on master (2:1, slow attack, 100ms release)

8. Configure the audio bus layout

Real audio in Summer Engine needs a bus layout. A new project starts with only the Master bus, so configure the remaining buses now.

Preferred (Summer MCP): Buses are typically in default_bus_layout.tres. Modify project setting:

summer_project_setting(key="audio/buses/default_bus_layout", value="res://audio/default_bus_layout.tres")

Ask the user:

May I create audio/default_bus_layout.tres with these buses: Master / Music / SFX / UI / Voice / Ambient / Reverb_Small / Reverb_Large / Reverb_Outdoor?

On yes, write the .tres directly via Write (it is the engine's plain-text resource format):

Write audio/default_bus_layout.tres

Fallback (no MCP): Print the .tres for the user to paste, plus the project.godot line.

9. Save the audio bible

Always write the bible. Path: .summer/audio-bible.md.

markdown
# Audio Bible — <Game Name>

**References:** <3 specific tracks / games / films>
**Style:** <one of the 8 above>
**Tempo / key:** 80–110 BPM, D minor, 4/4

## SFX Vocabulary
| Class | Character | Examples | DON'T |
|---|---|---|---|
| UI | Soft, woody, 60-100ms | hover, click | bell tones |
| Pickup positive | Warm major-third, 200-400ms | coin, heal | pure sine |
| ... | ... | ... | ... |

## Dynamic Music Plan
**Model:** Layered stems
**States:** CALM / TENSION / COMBAT / VICTORY
**Transitions:** ...
**Rules:** ...

## Spatial + Mix
- 3D attenuation: inverse-square, 25m
- Reverb buses: small / large / outdoor
- HRTF: ON
- Master ceiling: -1 dBTP
- Mix bus: Music -8 dB, SFX 0 dB, UI -3 dB
- Footstep rolloff: 4m player / 10m NPC

## Bus Layout
Master → Music, SFX (→ Reverb_Small / Large / Outdoor), UI, Voice, Ambient

Ask:

May I create .summer/audio-bible.md with this bible?

10. Optional — generate first SFX seeds

If the user wants to start generating, propose 3-5 SFX seeds aligned with the vocabulary. These are starting points, not finals. Generate them with the sound-effect skill: the first sound effect in a chat starts with its question (three versions from three models, picked by ear in Summer Studio), and its prompt shape is the one that tested well.

assets.generate_audio({ capability: "sound_effects", text: "Foley one-shot: a fingertip taps a small wooden button, soft woody click, no tail, close-mic, dry, no reverb, no music, no voices.", durationSeconds: 0.5 })
assets.generate_audio({ capability: "sound_effects", text: "Foley one-shot: a small gold coin drops onto a wooden table, bright metallic ting with a short ring, close-mic, dry, no music, no voices.", durationSeconds: 0.5 })
assets.generate_audio({ capability: "sound_effects", text: "Foley one-shot: a fist hits a padded leather vest, low-mid thump, no tail, close-mic, dry, no reverb, no music, no voices.", durationSeconds: 0.5 })

capability is required; sound_effects reads from text (not prompt), and the length argument is durationSeconds (not duration) with a floor of 0.5 s (1 s for Stable Audio 3 SFX). Leave out modelId so the person's approved default model runs. Ask for the character you want inside the text and let the player only use the front of the clip — you cannot request 80 ms.

Ask first:

May I generate first-pass SFX seeds for UI, pickup, and damage classes? They're starting points, not finals.

11. Route to the next step

End with:

Bible saved. Next:

  • /design-mechanic to wire SFX cues into specific mechanics.
  • /game-feel if you haven't paired audio with visual effect cues yet.
  • Author or generate the calm music track first; it's the longest-running and sets the room tone.

Common mistakes

MistakeFix
Single reference for musicThree minimum. Listen for the common element across them.
"Library" instead of "vocabulary"A bag of free SFX with no shared character feels random. Eight named classes with character notes.
Same SFX for all eventsIf pickup-coin and pickup-key sound identical, the player learns nothing from audio.
No dynamic music planStatic loops bore in 90s. Pick a model (layered / vertical / horizontal / procedural).
Music louder than SFXPlayers need SFX for gameplay; music is mood. Default mix is -8 dB music vs 0 dB SFX.
No reverb buses"Send" architecture (rooms route to a small/large/outdoor reverb bus) is mandatory or your mix sounds dry-then-wet inconsistently.
Pure sine "ding" for pickupReads as cheap mobile game. Use a major-third interval with character.
Same footstep sample for all surfacesPlayers hear footsteps constantly; material-aware steps carry much of the perceived audio quality.
No master ceiling / true peakAudio clips on console builds without -1 dBTP ceiling. Always set it.
Assuming SetResourceProperty fails silently on inline sub_resourcesIt does not — inline targets work, and the structural failures (node not found, property is not a resource, resource is null) are explicit errors. What is silent on current engines is a bad value shape: a JSON object instead of an engine literal string, a misspelled subProperty, or a wrong-typed value returns ok:true yet no-ops or coerces destructively. Pass class names and literal strings, and confirm in the saved .tres/.tscn, never from ok alone. See summer_docs_read id reference/mcp-tools-reference.

Collaborative protocol

This skill writes one bible (.summer/audio-bible.md), creates a bus layout (audio/default_bus_layout.tres), modifies project.godot, and may generate audio (metered). Always ask before writing or generating. See summer_docs_read id reference/collaborative-protocol.

Want a working starter?

No template — this is a workflow that produces the bible the rest of the project obeys. After the bible exists, the bus layout is set up and any subsequent audio work (footstep system, dynamic music) reads from it.

See also

  • summer_docs_read id reference/collaborative-protocol
  • summer_docs_read id reference/godot-version — Summer compatibility and version-sensitive audio API notes
  • summer_docs_read id reference/mcp-tools-reference
  • brainstorm-game — produces the brief that anchors the bible
  • art-direction — the visual counterpart; audio must rhyme with it
  • adaptive-music — implements the dynamic music plan
  • sound-effect, music-track, ambient-bed, voice-line — generate the audio the bible describes

From a shell

No Summer MCP tools in this session? Run each one as a command; when summer --version is missing or below 4.0, use npx -y summer-engine@latest in place of summer. Fields are flags in kebab-case (--scene-path res://main.tscn) or one --input '<json>', and summer ops describe <op> prints the exact input. The summer-cli skill has the rest.

In this skillCommand
summer_assets_generate_audiosummer assets generate-audio
summer_docs_readsummer docs read
summer_add_nodesummer engine add-node
summer_get_project_contextsummer engine get-project-context
summer_get_scene_treesummer engine get-scene-tree
summer_inspect_nodesummer engine inspect-node
summer_project_settingsummer engine project-setting

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 Audio Direction AI skill do?

Define the game's sound (music style, SFX vocabulary, dynamic music) as .summer/audio-bible.md.

Why use Audio Direction on TypingMind?

Because you install it once and use it with any model. Audio Direction 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 Audio Direction in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/SummerEngine/summer/tree/main/library/skills/audio-direction. 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 Audio Direction?

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 Audio Direction?

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

Is the Audio Direction AI skill free?

Yes. It is published on GitHub by SummerEngine 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 👇