gauntlet-loop

gauntlet-loop is a skill for Claude Code, Codex from duolahypercho/gauntlet-loop. It costs 128 tokens per session (770 once invoked), scanned A, original, MIT.

A prompt-driven workflow for building a video game against a named reference, such as a first-person shooter or roguelike. It uses repeated building, critique, and comparison until the person running it stops.

In plain words
What is it for?
Prototyping games, improving them through critique, and producing 2D or 3D game art with image-generation tools or Blender.
Why use it?
It provides a repeatable way to turn a broad game idea into an iterative development process without creating a separate scoring system.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/duolahypercho/gauntlet-loop/gauntlet-loop
Any agent
npx skills add duolahypercho/gauntlet-loop --skill gauntlet-loop
Clone the repo
git clone --depth 1 https://github.com/duolahypercho/gauntlet-loop

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for gauntlet-loop

README.md
[![agentmods](https://agentmods.dev/badge/skills/duolahypercho/gauntlet-loop/gauntlet-loop.svg)](https://agentmods.dev/skills/duolahypercho/gauntlet-loop/gauntlet-loop)
Your own site
<a href="https://agentmods.dev/skills/duolahypercho/gauntlet-loop/gauntlet-loop"><img src="https://agentmods.dev/badge/skills/duolahypercho/gauntlet-loop/gauntlet-loop.svg" alt="Measured on agentmods" height="20"></a>
Per session 128 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 770 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00128 $0.00770
Opus 5 $0.00064 $0.00385
Sonnet 5 $0.00026 $0.00154
Haiku 4.5 $0.00013 $0.00077

Measured 4d ago against content hash 18e89d3c422b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

gauntlet-loop 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.

skills/gauntlet-loop/SKILL.md · 57 lines

How it starts

The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Gauntlet Loop

This is a game skill. Pure prompt. No functions.

On invoke you are building a game to the level of a named reference game. Fill the three-paragraph aim prompt, then run it. Do not paste it for the user. Do not build a capture suite, state machine, or scoring framework around it. The prompt is the method.

  1. Read AGENTS.md — the prompt + game defaults + what not to invent.
  2. Read the harness overlay for loop verbs only:
  3. Infer nouns from $ARGUMENTS / the message. Default domain = game. One question max if THING is missing.
  4. Fill → execute on the game → keep going until the human stops you.
/gauntlet-loop a COD-style FPS in ThreeJS
/gauntlet-loop Hades-like roguelike in Godot
/gauntlet-loop Brotato-like arena survivor in Godot

Art for the game (not instead of the game)

Need Tool Doc
Sprites, textures, UI pixels, concepts Image gen ASSETS.md
2D hero/monsters (Brotato-like) Image gen sprite factory (style lock → hero → batch → import) ASSETS.md
Real meshes, GLB/FBX, 3D props/weapons Blender MCP BLENDER_MCP.md + ASSETS.md

Critic grades in-game frames vs the real reference game. Art tools only close asset gaps — then wire into the playable build. For 2D sprite games, characters/monsters are image-gen sprites with one locked art language — not Blender, not procedural ovals.

Do not

  • Treat this as a generic site/deck skill — games first
  • Dump the prompt and stop — you run it
  • Invent tools, harnesses, contracts, or round machinery
  • Soften the critic / lower the reference / invent a stop condition
  • End after one cycle and ask whether to continue
  • Mix harness verbs (/loop on Codex, /goal on Claude, ultracode on Codex)
  • Lag the game out with capture farms / headless engine spam — the game stays playable
  • Replace the game loop with endless image-gen or Blender-only rounds

Read the full file on GitHub · 57 lines

Files

What ships with it

6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 4d ago First seen · 57 lines · 128 tokens per session scan A 18e89d3c422b

Subscribe to this mod's changes

gauntlet-loop is a skill published in the GitHub repository duolahypercho/gauntlet-loop (150 stars, last pushed 1mo ago), licensed MIT. It adds 128 tokens to every session and 770 once invoked, about $0.0006 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 2026-08-30.