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/vchelaru/flatredball2/content-boundarynpx skills add vchelaru/FlatRedBall2 --skill content-boundarygit clone --depth 1 https://github.com/vchelaru/FlatRedBall2Wrote 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/vchelaru/flatredball2/content-boundary)<a href="https://agentmods.dev/skills/vchelaru/flatredball2/content-boundary"><img src="https://agentmods.dev/badge/skills/vchelaru/flatredball2/content-boundary.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 | $0.00065 | $0.02334 |
| Opus 5 | $0.00032 | $0.01167 |
| Sonnet 5 | $0.00013 | $0.00467 |
| Haiku 4.5 | $0.00006 | $0.00233 |
Grade A, and why
content-boundary 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 yesterday.
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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The AI / Human Content Boundary
FlatRedBall2 assumes a soft split of labor between AI and human. The split exists because AI has hard limits on a few things, and hiding those limits behind "AI does everything" produces worse games than embracing the split.
What AI Produces
- Code and structure — entities, screens, factories, collision wiring, state machines, input handling.
- Placeholders and scaffolding — valid-but-minimal TMX files, flat Gum screens, default coefficients, shape-based "programmer art" in place of sprites.
- Logic and integration — loading assets by known path, wiring coefficients from JSON, responding to collision events.
What the Human Produces
- Raster art — PNG sprites, backgrounds, UI art. AI cannot create these.
- Level design and placement — where platforms go, where enemies spawn, pacing, difficulty curve. AI cannot see a rendered level or play it to judge flow.
- UI composition — where controls sit on screen, visual hierarchy, typography. AI cannot see the rendered result.
- Feel tuning — jump height, run speed, friction, drag, attack timing. AI cannot feel gameplay.
AI and human can both edit code when needed, but the asymmetry is real: AI writing code is fast and reliable; AI composing art or tuning feel is slow and unreliable. Design around that.
Engine Design Implication — Externalize What the Human Tunes
When designing or reviewing an engine API, ask: will a human want to tune this without recompiling?
- Yes → the API must accept externalized data (JSON, TMX, .gumx, .achx). Example:
PlatformerValuesare consumed from JSON at runtime so designers can iterate in a text editor. - No → code-only is fine.
This is the lens behind decisions like JSON-driven platformer coefficients, TMX-driven level geometry, and .gumx-driven UI layouts. Avoid hardcoding anything a designer would reasonably want to tune by hand.
Operational Rule — Always Scaffold the Placeholder
When a game task adds a new piece of content, AI must create a placeholder file rather than hardcoding the content in C#. After scaffolding, tell the user which file to open in which tool.
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.
- yesterday First seen · 129 lines · 65 tokens per session scan A 97745864e94e
content-boundary is a skill published in the GitHub repository vchelaru/FlatRedBall2 (14 stars, last pushed today), licensed MIT. It adds 65 tokens to every session and 2,334 once invoked, about $0.0003 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 2026-09-04.
Other skills, from other repositories
gameobject-component-destroy
Destroy one or more Components from a target GameObject. Missing (null) components are skipped — they cannot be destroyed. Use 'gameobject-find' and 'gameobject-component-get' to identify the components first.
unity-version-split
Split a C# file into Unity 6.5+ and pre-Unity 6.5 variants. Use when a file needs different implementations for different Unity versions due to API changes (e.g., EntityId vs int, GetEntityId vs GetInstanceID).
playtest-report
Generates a structured playtest report template or analyzes existing playtest notes into a structured format. Use this to standardize playtest feedback collection and analysis.
unity-manual-component
Manually add, configure, reorder, and copy components on GameObjects using Unity Editor UI. For one-off Inspector workflows that do not need REST automation.
godot-signals-groups
Build event-driven, decoupled Godot 4.7 gameplay with signals and node groups: declare and emit custom signals, connect with Callables (incl. bind/one-shot), and broadcast to many nodes via groups and callgroup. Use when wiring node communication in a Godot project, replacing tight references with signals…
threejs-exposure-color-grading
Build a measured exposure and grading path in Three.js. Use for a 64x36 encoded luminance meter, asynchronous readback, weighted log-average exposure, asymmetric adaptation, single tone-map ownership, and a generated 32-cube post-tone-map LUT.