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 skills/frabcd/codex-ai-game-studio/setup-windows-editionnpx skills add frabcd/codex-ai-game-studio --skill setup-windows-editiongit clone --depth 1 https://github.com/frabcd/codex-ai-game-studioWrote 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/frabcd/codex-ai-game-studio/setup-windows-edition)<a href="https://agentmods.dev/skills/frabcd/codex-ai-game-studio/setup-windows-edition"><img src="https://agentmods.dev/badge/skills/frabcd/codex-ai-game-studio/setup-windows-edition.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 5.1 | $0.00088 | $0.00842 |
| Opus 5 | $0.00044 | $0.00421 |
| Sonnet 5 | $0.00018 | $0.00168 |
| Haiku 4.5 | $0.00009 | $0.00084 |
Grade A, and why
setup-windows-edition scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Set up the Windows edition
Keep discovery read-only and prefer native Windows execution. Do not launch an editor, trust a hook, install a package, download a model, enable an MCP server, write project configuration, or cross the WSL boundary during discovery.
-
Resolve
PLUGIN_ROOTto this plugin's installed root, resolve the project root, preserve its current Git state, and run the native launcher:& "${PLUGIN_ROOT}\scripts\ai-game-studio-windows.ps1" doctor --project <root>Verify native Windows versus WSL,
amd64versusarm64, PowerShell availability,winget/Chocolatey/Scoop presence, installed engines and DCC tools, active MCP host selections, GPU vendor, CUDA/DirectML/CPU/WARP options, disk space, network needs, and credential variable names without reading secret values. -
Read the bundled descriptor at
${PLUGIN_ROOT}/editions/windows.json. Reject the Windows edition if the target is not native Windows. Treat WSL as a disclosed fallback only; do not use it for GUI editor control, project files on a cross-boundary filesystem, or a capability that has a verified native route. -
Select one route in this order:
- Adapt portable source and launchers to native PowerShell or standard-library Python when the upstream license, architecture, and tests support it.
- Select a verified capability-equivalent Windows alternative and compare capability, output quality, cost, performance, privacy, license, and limitations.
- Offer a hosted or manual handoff only when neither native route works.
Never describe a macOS application bundle, POSIX executable, Apple-Silicon binary, or Metal implementation as Windows-compatible. Every substitution requires confirmation.
-
Create the proposal:
& "${PLUGIN_ROOT}\scripts\ai-game-studio-windows.ps1" plan --project <root> --output <plan.json>Use a user-approved temporary plan path outside the project for preview-only work. Present the detected environment, exact actions, immutable pins, downloads, licenses, permissions, conflicts, expected artifacts, backups, rollback operations, expiry, and SHA-256 digest. State which checks are evidence and which capabilities remain unverified.
-
Stop and wait for the user to repeat the displayed digest verbatim. Do not interpret general approval as digest confirmation. Apply only the unexpired, unchanged proposal:
& "${PLUGIN_ROOT}\scripts\ai-game-studio-windows.ps1" apply --project <root> --plan <plan.json> --confirmed-digest <digest> -
Run the descriptor health checks after apply. Confirm exact executable paths and hashes, one active MCP server per host application, editor/version compatibility, GPU backend availability, project parse/import status, and transaction-owned files. Do not launch editors or generated content unless those actions were in the confirmed proposal.
-
Report the transaction ID, evidence paths, limitations, and rollback route. Generate disable or rollback as a new proposal and require a new digest:
& "${PLUGIN_ROOT}\scripts\ai-game-studio-windows.ps1" disable --project <root> --output <disable-plan.json> & "${PLUGIN_ROOT}\scripts\ai-game-studio-windows.ps1" rollback <transaction-id> --project <root> --output <rollback-plan.json>
For a preview request, stop after presenting the plan. Do not write .ai-game-studio/project.json, .ai-game-studio/lock.json, editor configuration, MCP configuration, or package state.
What ships with it
1 file 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.
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.
- 5d ago First seen · 53 lines · 88 tokens per session scan A 98ecc08d458e
setup-windows-edition is a skill published in the GitHub repository frabcd/codex-ai-game-studio (9 stars, last pushed 5d ago), licensed MIT. It adds 88 tokens to every session and 842 once invoked, about $0.0004 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 skills, from other repositories
3d_object
Judge whether a generated mesh is fit to ship in a browser game, from the same rendered sheet the orientation review uses.
motion
How an agent turns a character mesh into a usable animated FBX — and how to judge whether the result is shippable.
audio
Use this Skill when a game plan requires dialogue, voice lines, sound effects, foley, ambience, or other offline WAV assets. This is asset generation, not a runtime audio-playback contract; use the selected engine context after an asset has passed review.
novel-to-game
Turn a novel into a fully playable game on the selected target platform. Orchestrates the whole adaptation pipeline — requirements intake, gameable deconstruction, concept selection, world and visual design, target-runtime build, and evidence-based QA — for a novel in any language. Use for novel to game, story to…
game-build
Build a risk-matched whitebox or the approved production game for its target runtime. Turn GAMEDESIGN, and ARTDIRECTION when production begins, into a minimal BUILDBRIEF and a runnable candidate that can be iterated with replayable evidence. Use for prototype the riskiest design question, implement the approved game…
game-concept
Design game concepts from a novel. From SOURCEBIBLE and PRODUCTBRIEF, generate three genuinely different directions on the dimensions still unlocked, then pick the most worthwhile playable prototype using hard vetoes and explicit trade-offs. Use for what game should this novel become, compare game concepts, choose a…