Estella is a 2D and 3D game engine with a TypeScript interface, a C++/WebAssembly core, and a visual editor that also acts as an MCP server. It helps developers create, test, and export games for the web, desktop, mini-game platforms, playable ads, and native mobile devices. Its catalogue add-ons let coding agents operate the editor through the same tool surface used by its interface.
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 esengine/estella --skill estella-debugginggit clone --depth 1 https://github.com/esengine/estellaWrote 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/esengine/estella/estella-debugging)<a href="https://agentmods.dev/skills/esengine/estella/estella-debugging"><img src="https://agentmods.dev/badge/skills/esengine/estella/estella-debugging/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/esengine/estella/estella-debugging"><img src="https://agentmods.dev/badge/skills/esengine/estella/estella-debugging.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00055 | $0.00850 |
| Opus 5.5 | $0.00022 | $0.00340 |
| Sonnet 5.5 | $0.00011 | $0.00170 |
| Haiku 4.5 | $0.00006 | $0.00085 |
Grade A, and why
estella-debugging 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 today.
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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Finding why it fails
Start from verify (see estella-workflow): it names the failed check and its findings. Fix the first finding of the
first failed check before anything else; later ones often follow from it. Then look in this order.
The game did not start (play failed)
get_logs { tail: 50 } has the first error. A script that throws when the game starts, a component registered twice, an
asset that will not load. Fix it, then verify again: nothing inside the game was checked while it could not start.
An error while it ran (errors failed)
The finding's detail has the stack. read_project_file the file and line it names. A script reading a component the
entity does not have is the usual one: world.get throws for a missing component, world.tryGet answers null.
The scene is incomplete (scene failed)
get_diagnostics lists each problem by entity, component and field: a required field left empty (an AudioSource with no
clip), a reference to an asset that does not exist, an asset that loaded with part of it missing. list_assets shows what
does exist.
Nothing is drawn (look failed, or the screenshot is empty)
- Is there a camera?
find_entities { component: "Camera" }in the running game; it must be active. - Does it see the content?
inspect_entitythe camera and the thing that should show: an orthographic camera showsorthoSizeworld units above and below its position, so a sprite at y 2000 under an orthoSize of 540 is off screen. - Is the content drawable? A
Spriteneeds a size, aMeshRenderera mesh and a material; the entity and its parents must be active. A Sprite with no texture draws a tinted quad, not nothing. - UI: a screen
Canvasdraws on top of every camera; a node with no size ordisplay: nonedraws nothing. capture_framelists every draw of the frame with the entities it came from: what is not in it was not submitted.
Input does nothing
play_input sends canvas-relative screen pixels, y down, and key codes as KeyboardEvent.code (Space, KeyA,
ArrowLeft). After a press, step { frames: 1 } or more: the press lasts one frame. A UI control under the pointer takes
the event first. inspect_entity the controlled entity before and after: if its input component changed and its position
did not, the problem is movement (physics, a script), not input.
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.
- today First seen · 60 lines · 55 tokens per session scan A 4ee53a221884
estella-debugging is a skill published in the GitHub repository esengine/estella (1,174 stars, last pushed today), licensed Apache-2.0. It adds 55 tokens to every session and 850 once invoked, about $0.0002 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 2026-11-19.
estella-debugging compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| estella-debuggingesengine/estella | Skill | 1.2k | today | Apache-2.0 | A | 55 | Claude Code, Codex |
| realvirtual-web-commissioninggame4automation/realvirtual-WEB | Skill | 194 | 12d ago | AGPL-3.0 | A | 48 | Claude Code, Codex |
| migrate-oxlintoxc-project/oxc | Skill | 24k | today | MIT | A | 44 | Claude Code, Codex |
| cli-wiringgithub/gh-aw | Skill | 5.7k | today | MIT | A | 0 | Claude Code, Codex |
| adversarial-revieweremdash-cms/emdash | Skill | 14k | today | MIT | A | 72 | Claude Code, Codex |
| deep-investigationvercel/eve | Skill | 6.0k | today | Apache-2.0 | A | 28 | Claude Code, Codex |
| investigatevercel/eve | Skill | 6.0k | today | Apache-2.0 | A | 25 | Claude Code, Codex |