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 agentmods add instructions/shellishack/3d-web-game-dev-skills/agents-mdgit clone --depth 1 https://github.com/Shellishack/3d-web-game-dev-skillsWrote 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/instructions/shellishack/3d-web-game-dev-skills/agents-md)<a href="https://agentmods.dev/instructions/shellishack/3d-web-game-dev-skills/agents-md"><img src="https://agentmods.dev/badge/instructions/shellishack/3d-web-game-dev-skills/agents-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00528 | $0.00528 |
| Opus 5 | $0.00264 | $0.00264 |
| Sonnet 5 | $0.00106 | $0.00106 |
| Haiku 4.5 | $0.00053 | $0.00053 |
Grade A, and why
3d-web-game-dev-skills AGENTS.md 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 3d 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Instructions
This repository contains shareable Codex skills for building game projects. Treat any external game repository used as inspiration as raw research material, not as content to preserve.
Summarization Standard
When extracting patterns from an existing project, summarize at the level another agent needs to reproduce the architecture:
- Describe folder responsibilities, module boundaries, runtime responsibilities, and data flow.
- Name generic roles such as
game/,ui/,bridge/,assets/,controllers/,npc/, andscenes/. - Capture reusable contracts and rules, such as where simulation state belongs, where React UI state belongs, and when a bridge context is appropriate.
- List package families and why they are used, such as renderer, UI framework, build tool, styling, desktop wrapper, icon library, or animation library.
- Convert concrete filenames to reusable examples when a name is domain-specific.
Prefer summaries that help an agent scaffold a similar project from scratch. Avoid summaries that read like an inventory of a specific product.
Generalization Rules
Before adding content to a skill, convert source-specific details into reusable guidance:
- Replace product names, repository paths, character names, place names, story terms, and domain-specific labels with generic terms.
- Replace private game data with structural descriptions, type shapes, or placeholder examples.
- Replace source-specific component names with generic component roles when the original name is not itself a reusable convention.
- Keep exact package names, framework names, commands, and file extensions when they are part of the reusable technical pattern.
- Keep small reusable code assets only when they are generic, self-contained, and not tied to a private theme or product.
Skill Content
Skills in this repository should contain only what future agents need:
SKILL.mdfor concise workflow and decision guidance.references/for longer reusable architecture notes or package summaries.assets/for generic templates or components that can be copied into new projects.scripts/only when a repeatable operation benefits from deterministic tooling.
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.
- 3d ago First seen · 47 lines · 528 tokens per session scan A 8f0a1037d90b
3d-web-game-dev-skills AGENTS.md is an instructions file published in the GitHub repository Shellishack/3d-web-game-dev-skills (5 stars, last pushed 2mo ago), licensed MIT. It adds 528 tokens to every session, about $0.0026 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 2026-08-31.
Other instructions, from other repositories
kart-royale CLAUDE.md
Instructions for ryancampbell/kart-royale, covering working on this repo, commands, the contract, traps that have already cost a round and verifying a change.
babylonjs-webpack-es6 AGENTS.md
Instructions for RaananW/babylonjs-webpack-es6, covering agents.md, what this project is, layout, how to add a scene (the only supported way) and babylon.js import rules (most common source of bugs).
babylonjs-webpack-es6 copilot-instructions.md
Instructions for RaananW/babylonjs-webpack-es6: This repository is a Babylon.js + TypeScript + webpack (ES6 modules) starter template.
warbell AGENTS.md
AGENTS.md instructions for miskibin/warbell, covering agents.md and cursor cloud specific instructions.
warbell CLAUDE.md
Claude Code instructions for miskibin/warbell, covering claude.md, what this is, models / subagents, commands and git / agent workflow rules (non-negotiable).
threejs-agent-starter AGENTS.md
Instructions for heagandev/threejs-agent-starter, covering agent rules — three.js game dev starter pack, stack, architecture rules, three.js rules and code style.