Asset Strategy logo

Asset Strategy

Organization
SummerEngine
asset-strategy

Route any 'I need a [thing]' asset request to the right skill: 2D, 3D, audio, video, VFX or animation.

Overview

PublisherSummerEngine
Repositorysummer
Skill nameasset-strategy
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 Asset Strategy 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/asset-strategy
  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/asset-strategy .claude/skills/asset-strategy
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Asset Strategy 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 Asset Strategy 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 Asset Strategy 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.

Asset Strategy — Meta-Router for Asset Creation

Classify the user's request, ask at most one clarifying question, then dispatch to the right specialist via the Skill tool. Don't generate anything yourself — every media-type/asset-class combination has a dedicated specialist that owns the prompt patterns, polycount targets, gates, and fallbacks. Your job is the routing decision, not the work.

Step 1 — Classify by media type

Signal in user requestContinue at
image, art, concept, portrait, icon, sprite, texture, skybox, HUD2D table
model, mesh, 3D, prop, character, rigged, vehicle, tree, level kit3D table
sound, SFX, music, ambient, voice, VOAudio table
cutscene, trailer, loop video, animated backgroundVideo table
animation, idle, walk, attack, retarget, blend tree, IK, lipsyncAnimation table
fire, muzzle flash, smoke, lightning, shake, hit-stop, particlesVFX table

If the request spans rows ("a witch with idle and a fire spell"), classify the first asset and dispatch — the specialist hands off when it's done.

Step 2 — Disambiguate (one question max)

Ask the exact phrase before dispatching. Skipping the question is the most common routing mistake.

AmbiguityDisambiguation question (verbatim)Routes
Concept art vs character portrait"Quick check — exploring the look (3-4 variants for art direction) or generating a final character portrait for dialogue UI?"concept-art (exploring) / character-portrait (final)
Prop vs character model"Static object (sword, chair, lantern) or rigged character that needs animation?"prop-model / character-model
Tile vs sprite sheet"One seamless tile that repeats, or a frame grid for an animated sprite?"tileable-texture / sprite-sheet
SFX vs music"One-shot effect or a longer track (loop, theme, ambient bed)?"sound-effect / music-track (or ambient-bed)
Generate vs retarget motion"Generate a new clip on this character, or apply an existing library to a new character?"generate-motion / retarget
VFX recipe vs game-feel"Particle/shader effect in the world or screen-feel (camera shake, hit-stop)?"vfx-<name> / game-feel

Step 3 — Dispatch

Invoke the specialist via the Skill tool. Don't paraphrase — let it run. Pass the user's original phrasing through.

Routing tables

2D assets

RequestSkill
Concept / mood board / variantsconcept-art
Polished portrait / NPC bustcharacter-portrait
Pixel sprite / retro iconpixel-art
HUD / button / inventory frameui-graphics
Wall / floor / seamless groundtileable-texture
Run cycle / spritesheet gridsprite-sheet
360 sky / panoramic backgroundskybox-panorama

3D assets

RequestSkill
Sword / chair / barrel / static objectprop-model
Player / NPC / enemy / boss / rigged humanoidcharacter-model (generated mesh + auto-rig)
Modular dungeon / building blocksenvironment-kit
Car / spaceship / hard-surface vehiclevehicle-model
Tree / rock / mushroom / foliageorganic-model

Polycount targets (hero 8k–15k, grunt 3k–6k, mobile 1k–3k, cinematic 20k–40k) live in each specialist.

Audio

RequestSkill
Sword clash / footstep / UI clicksound-effect
Theme / level music / boss trackmusic-track
Forest ambience / cave droneambient-bed
NPC line / VO / spoken dialoguevoice-line
Music that reacts to combatadaptive-music
What should this game sound likeaudio-direction

Animation

RequestSkill
Idle / walk / run / attack — new clipgenerate-motion
Custom action or non-humanoid rig (flower, creature, prop) — clip from texttext-to-motion
Apply library to a different characterretarget
Blend tree / state machineanimation-tree
Look-at / IK / foot placementprocedural-animation
Lipsync / facial blendshapesfacial-and-lipsync

Video

RequestSkill
Intro cutscene / story sequencecinematic-cutscene
Steam trailer / marketing cliptrailer-shot
Animated menu background / loopanimated-loop

Visual effects

VFX is code, not generative — recipes are shader + GDScript + node setup, not image generation.

RequestSkill
Fire / torch / campfirevfx-fire
Muzzle flash / gunshotvfx-muzzle-flash
Water ripple / splashvfx-water-ripple
Lightning / electric arcvfx-lightning
Dissolve / disintegratevfx-dissolve
Smoke / dust puffvfx-smoke
Hit spark / impact flashvfx-hit-spark
Magic glow / auravfx-magic-glow
Camera shake / hit-stopgame-feel

How the specialists call Summer

The specialists name Summer's asset work by operation. A call written area.action({...}) is one operation: use the tool summer_area_action when your client lists it, else summer_call({ op: "area.action", input: {...} }). summer_search finds any operation and its exact input.

  • Find and read: assets.search({ query, scope }) (scope: library (default), mine for their own, all), then assets.get({ assetId }) for the file, downloadUrl and viewerUrl. Reuse before generating.
  • Generate: assets.generate_image, assets.generate_3d, assets.generate_audio, assets.generate_video, assets.generate_motion, assets.generate_sfx_choices (then assets.choices for the person's picks). They spend the person's credits; dryRun: true returns the estimate (credits, balance, enough) and makes nothing.
  • Wait: 3D, video and animation answer at once with a job; jobs.wait({ jobId, timeoutSeconds: 50 }) until finished, and the finished job's result holds the asset ids.
  • Save on this machine: with the local Summer MCP or the CLI, out (a folder or a file path) on a generate call, jobs.wait or assets.get saves the file there, and the answer adds localPath.
  • Edit their assets: assets.remove_background, assets.slice_sheet, assets.update (title, description, tags, private or team), assets.make_public (lists one for everyone in Summer's asset store), assets.delete (deletes for good; asks the person unless they let their agent delete, and then nothing is deleted until they confirm).
  • Permissions: The person decides once what their agent may do: publishing (including assets.make_public), spending and building run at once by default, and deleting for good asks. An operation whose group they have not granted answers needs_approval with a link: give the link to the person, then jobs.wait on the apr: job.
  • Upload an image from disk: assets.upload_begin, then assets.upload_complete.
  • One-shot makers: makers.<tool> (for example makers.npc_voice, makers.trailer_forge); summer_search lists them.

When NOT to use this skill

  • The user already named the specific skill — go straight there.
  • The user is debugging or running the game — route to debug or play.
  • The asset is an existing Summer asset ID — call assets.get({ assetId }), then summer_import_asset_by_id.
  • The asset is being imported from a non-Summer URL (.glb / .wav from Sketchfab, Quaternius, AmbientCG) — call summer_import_from_url directly.
  • The user wants to plan the whole game's asset list — brainstorm-game first, then this router on each item.

Anti-patterns

  • Generating yourself instead of dispatching. This file holds zero prompt patterns by design. If you're writing an assets.generate_image call from here, stop and invoke the specialist.
  • Skipping the disambiguation question. Sends users to the wrong specialist and burns a paid generation. The verbatim phrasing in Step 2 is required.
  • Routing concept-art to character-portrait (or the reverse). Different goals: concept-art produces 4 divergent variants for art direction; character-portrait produces 1 polished bust for dialogue UI.
  • Inventing skill names. Only dispatch to skills listed above. If nothing fits, ask the user to clarify.
  • Routing rigged-humanoid requests to prop-model. Anything that needs animation always goes to character-model.

Without Summer Engine

In a session without Summer Engine tools (the assets profile; the summer-assets skill has the whole loop), the routing tables still pick the skill. Leave out every step that imports (summer_import_asset_by_id, summer_import_from_url): deliver the file with out instead (the summer-assets skill says how).

Fallback (no MCP)

The router still works without MCP — classification is text-only. It hands off to the specialist, which surfaces its own no-MCP fallback (typically: run the same generation in the Summer dashboard, then summer_import_from_url once MCP is back).

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
assets.choicessummer assets choices
assets.deletesummer assets delete
assets.generate_3dsummer assets generate-3d
assets.generate_audiosummer assets generate-audio
assets.generate_imagesummer assets generate-image
assets.generate_motionsummer assets generate-motion
assets.generate_sfx_choicessummer assets generate-sfx-choices
assets.generate_videosummer assets generate-video
assets.get, summer_assets_getsummer assets get
assets.remove_backgroundsummer assets remove-background
assets.search, summer_assets_searchsummer assets search
assets.slice_sheetsummer assets slice-sheet
assets.updatesummer assets update
assets.upload_beginsummer assets upload-begin
assets.upload_completesummer assets upload-complete
summer_import_asset_by_idsummer engine import-asset-by-id
summer_import_from_urlsummer engine import-from-url
jobs.waitsummer jobs wait
makers.npc_voicesummer makers npc-voice
makers.trailer_forgesummer makers trailer-forge
summer_searchsummer ops search "<what you want>"

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 Asset Strategy AI skill do?

Route any 'I need a [thing]' asset request to the right skill: 2D, 3D, audio, video, VFX or animation.

Why use Asset Strategy on TypingMind?

Because you install it once and use it with any model. Asset Strategy 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 Asset Strategy in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/SummerEngine/summer/tree/main/library/skills/asset-strategy. 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 Asset Strategy?

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 Asset Strategy?

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

Is the Asset Strategy 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 👇