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 kyh/vibedgames --skill level-designgit clone --depth 1 https://github.com/kyh/vibedgamesWrote 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/kyh/vibedgames/level-design)<a href="https://agentmods.dev/skills/kyh/vibedgames/level-design"><img src="https://agentmods.dev/badge/skills/kyh/vibedgames/level-design/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kyh/vibedgames/level-design"><img src="https://agentmods.dev/badge/skills/kyh/vibedgames/level-design.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00031 | $0.01701 |
| Opus 5.5 | $0.00012 | $0.00680 |
| Sonnet 5.5 | $0.00006 | $0.00340 |
| Haiku 4.5 | $0.00003 | $0.00170 |
Grade A, and why
level-design 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 1mo 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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Level design
Levels are authored arguments: each one teaches an idea, tests it, twists it, and concludes. These patterns apply equally to hand-built tilemaps and code-generated levels — generators should emit the same grammar.
Core rules
- Four beats per level (kishōtenketsu — Hayashida, Super Mario 3D World): introduce the mechanic safely → develop with real stakes → twist (one variable flipped) → conclude with a capstone. One mechanic per level.
- The level is the tutorial. First exposure to anything new cannot kill: show the crusher cycling over a pit you can't fall into. Never explain in text what geometry can teach (Hayashida; Thorson).
- Pull with eyes, not arrows (the weenie — Rogers, via Disneyland): the goal or its marker visible from afar; leading lines (platform edges, coin trails) aimed along the critical path; one saturated accent color reserved for interactables; motion for anything that must not be missed.
- Experiential density (Valve's Half-Life cabal): a stimulus — enemy, pickup, set piece — every few seconds of distance traveled, never timers. Audit generated levels for dead stretches >3–4s of travel.
- Deaths are the player's fault: telegraph every threat (wind-up, audio cue, flash) and leave a readable escape. Surprise kills make players blame the game and quit (Valve).
- Sawtooth difficulty: rise ~4–5 levels/waves, drop deliberately, rise from a higher floor. Every 4th–5th level is a victory lap remixing known elements. Rest stops (hazard-free screen, checkpoint, secret) after peaks.
- Short rooms, instant retry for execution challenges: death costs <2s, respawn at room entry, no fade (Celeste).
- Two routes per room where possible: safe-slow vs risky-fast with a reward on the risky one (Thorson). Asymmetric layouts — mirror-symmetric rooms read artificial.
- Secrets reward attention, not pixel-hunting: signal with an anomaly — off-pattern tile, coin pointing into a wall.
- Cut, don't patch: a room that's still broken after 2 iterations gets deleted (Thorson; Valve).
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.
- 1mo ago Changed · -86 tokens per session 3e1ce49d7f0e
- 1mo ago First seen · 123 lines · 117 tokens per session scan A ac863b0d778f
level-design is a skill published in the GitHub repository kyh/vibedgames (61 stars, last pushed today), licensed MIT. It adds 31 tokens to every session and 1,701 once invoked, about $0.0001 per session on Opus 5.5. 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.
Other skills, from other repositories
retro-learn
Convert delivery findings into skill, eval, workflow, and documentation improvements.
system-design-case-catalog
Answer classic system design problems as constraint-to-solution sketches and coach interview practice: URL shortener, rate limiter, news feed, chat, notification, autocomplete, crawler, unique id. Use for interview practice or naming the closest known shape for a new problem.
system-design-interview-coaching
Facilitate timed system-design interview practice across backend, mobile/client, SDK/library, and full-stack surfaces. Use risk-driven probes, an adverse timeline, mechanism revision, one changed constraint, and the existing six-criterion evidence-based debrief.
supabase
Use when doing ANY task involving Supabase: Database, Auth, Edge Functions, Realtime, Storage, Vectors, Cron, Queues, supabase-js, @supabase/ssr, RLS, schema migrations, CLI, MCP server. Includes security checklist.
React Performance
Skill "React Performance" from databayt/hogwarts, covering react performance skill, memoization, code splitting and checklist.
migrate-radix-to-base
Migrates React projects and components from Radix UI to Base UI. Use when asked to migrate from radix, move to base-ui, convert radix primitives, or switch a shadcn project's base library. Handles single components ("migrate accordion") and whole projects.