Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add MRCalderon3D/everything-game-dev-code --skill ai-asset-generationgit clone --depth 1 https://github.com/MRCalderon3D/everything-game-dev-codeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mrcalderon3d/everything-game-dev-code/ai-asset-generation)<a href="https://agentmods.dev/skills/mrcalderon3d/everything-game-dev-code/ai-asset-generation"><img src="https://agentmods.dev/badge/skills/mrcalderon3d/everything-game-dev-code/ai-asset-generation.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 118 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00043 | $0.01944 |
| Opus 5 | $0.00022 | $0.00972 |
| Sonnet 5 | $0.00009 | $0.00389 |
| Haiku 4.5 | $0.00004 | $0.00194 |
Grade A, and why
ai-asset-generation scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 155 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Asset Generation
Purpose
Generate production-candidate game assets from text prompts through the generative
provider registry (manifests/asset-providers.json) so projects can move from
procedural placeholders to real content without leaving the scaffold — while
keeping generation engine-neutral, reproducible, provenance-tracked, and subject
to the same acceptance gates as authored content.
Optional Capability — Gate Before Use
AI generation is an optional layer, not a dependency. It is available only when the
active provider's API key is set in the environment (FAL_KEY for the default
fal.ai provider). The scaffold's asset workflow does not require it:
- Key present → this skill is available; the project may upgrade placeholders to generated content.
- Key absent → this skill is inert. Fall back to the default scaffold tooling
exactly as before the capability existed —
placeholder-asset-pipelineand the engine placeholder commands (/unity-placeholders,/godot-placeholders,/web-placeholders) for sprites, 3D primitives, and procedural audio. A project built entirely on placeholders is a fully valid, shippable-to-prototype outcome; generation is an enhancement, never a gate.
Detecting the gate is cheap: the generator (scripts/generate-assets.js) reads the
key from the environment and reports clearly when it is missing, pointing back to
the placeholder path. Never fabricate or hand-place assets to "stand in" for the API.
Resolution Order and Cost
- Prefer free before paid. For
image/skybox(the registry'snativeFirstCapabilities), if the running harness has a native image generator (e.g. Codex$imagegen), use it instead of the paid API — keep the same names, paths, and acceptance gates. Fall to the API only when no native generator exists and a key is set; fall to placeholders otherwise.model3d/sfx/music/speech/videohave no native path — API or placeholders only. - Confirm before spending. Every run prints an estimated cost; the generator
refuses runs at or above
confirmOverUsdwithout--yes. For a single asset the user explicitly asked for, dry-run, show the estimate, and proceed if cheap. For autonomous or batch flows, get a clear yes on the total before running. Iterate on the cheap default model; reserve the expensive alternative and video for finals. - Quality vs price tier. Ask the user once, before a batch, which tier to use —
budget,balanced, orpremium— and pass--quality(defaultbalanced). The registry'sbyQualitymap picks the model per capability (e.g. image: flux-schnell / nano-banana-2 / nano-banana-pro; video: Wan / Seedance / Veo). Budget can keep a whole video run under the cost gate; premium crosses it and needs--yes. Tier sets the spend; the gate prevents surprises.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 155 lines · 43 tokens per session scan A cd8303b2784d
ai-asset-generation is a skill published in the GitHub repository MRCalderon3D/everything-game-dev-code (81 stars, last pushed 18d ago), licensed MIT. It adds 43 tokens to every session and 1,944 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
asset-gen
Generate visual assets from text prompts: PNG images (Gemini / xAI Grok), GLB 3D models (Tripo3D), rigged biped characters, retargeted animations, and frame-by-frame animated sprites, plus background removal. Use whenever a game needs generated art.
art-generate
Generates placeholder .aseprite files from asset specs using the Aseprite MCP. Reads asset specs and art bible, creates sprites with correct dimensions/palette/layers, exports PNGs. Run after /asset-spec has produced specs and /art-bible exists.
cgs-art-bible
Use for art bible tasks that define visual identity, shape language, palette, camera, animation, UI style, and asset constraints; produce verification evidence, changed or proposed files, and handoff boundaries.
cgs-team-audio
Use for team audio tasks that coordinate audio design, implementation, asset readiness, mix targets, and QA for audio work; produce verification evidence, changed or proposed files, and handoff boundaries.
team-audio
Orchestrate audio team: audio-director + sound-designer + technical-artist + gameplay-programmer for full audio pipeline from direction to implementation.
team-narrative
Orchestrate the narrative team: coordinates narrative-director, writer, world-builder, and level-designer to create cohesive story content, world lore, and narrative-driven level design.