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 tettethu/VibeGame --skill artist-self-evolvegit clone --depth 1 https://github.com/tettethu/VibeGameWrote 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/tettethu/vibegame/artist-self-evolve)<a href="https://agentmods.dev/skills/tettethu/vibegame/artist-self-evolve"><img src="https://agentmods.dev/badge/skills/tettethu/vibegame/artist-self-evolve/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/tettethu/vibegame/artist-self-evolve"><img src="https://agentmods.dev/badge/skills/tettethu/vibegame/artist-self-evolve.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 5 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Rogue Agent · line 2 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
- high Rogue Agent · line 3 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
- high Rogue Agent · line 78 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
- high Rogue Agent · line 142 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
- high Rogue Agent · line 6 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
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 | $0.00062 | $0.02315 |
| Opus 5.5 | $0.00025 | $0.00926 |
| Sonnet 5.5 | $0.00012 | $0.00463 |
| Haiku 4.5 | $0.00006 | $0.00231 |
Grade A, and why
artist-self-evolve 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 — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Artist Self-Evolve
Learn, from a completed project, the prompt patterns and art rules that reliably produced its assets. The lessons are stored in two places:
skeletons/<slug>/art-pack.md— the sub-genre's empirical raw-asset inventory (prompt + layout + Hints per raw image). Direct write, full responsibility.spec/contracts/<x>.md#### Artistchapters — per-pattern asset workflows. You fill the chapter when lead drafts a new or revised contract.
Input
The lead's dispatch message provides:
game-slug— the project's sub-genre slug. All per-project outputs land underskeletons/<game-slug>/.
If game-slug is missing from the dispatch message, reply to the lead asking for it. Do not guess from .vibegame/GDD.md or project.json — the slug is the lead's decision, not yours.
Workflow
Phase 0: Find actually-used assets
Scan, in this order:
assets/manifest.json— every registered asset..vibegame/assets.md— asset notes and status..vibegame/GDD.md— what the design called for.- The game's runtime code — which manifest keys actually load at runtime.
The output of this phase is the empirical final-asset list — the manifest keys actually loaded at runtime, minus anything unused / replaced / stale. This list is internal to your run; you do not write it to a file. Use it as Phase 1's input.
Phase 1: Trace each final asset back to its raw image; recover prompts; write art-pack.md
A final asset rarely equals one raw vibegame art gen image output. The same raw image can produce multiple final assets — e.g. one decomposed sheet is cut into icon + slot + bar, three final manifest keys. Conversely, every raw image was produced by exactly one prompt invocation (one imagegen.jsonl row).
So the art-pack is keyed by raw image, not by final asset. Each ### <asset> heading represents one vibegame art gen image call — one prompt, one layout, one imagegen.jsonl row. The post-processing that turns a raw into one-or-many final assets is out of scope for art-pack; it belongs in the contract's #### Artist chapter (Phase 2).
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 First seen · 145 lines · 62 tokens per session scan A 5d369ef84843
artist-self-evolve is a skill published in the GitHub repository tettethu/VibeGame (283 stars, last pushed 14d ago), licensed Apache-2.0. It adds 62 tokens to every session and 2,315 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 2026-08-30.
Other skills, from other repositories
hearth-art
Give a Hearth game real art and sound — importing and slicing spritesheets, animations, procedural sprites and sounds, autonomous CC0 asset sourcing (Kenney, itch.io, OpenGameArt, Freesound, Google Fonts) with licensing rules, and pixel-art discipline (never stretch; read the art before using it). Use when the game…
hearth-build
Structure a Hearth game's world — scenes, entities, components, tilemaps and autotiling (surfaces must connect), collider/sprite alignment, prefabs, animation state machines, and input bindings. Use when creating or arranging what exists in a Hearth game (levels, objects, hierarchies); behavior scripts are…
hearth-feel
Make a Hearth game feel good, not just run — the juice stack (hit-stop, screen shake, flash, particle bursts, layered sound), tween easing, camera effects, anticipation/recovery animation idioms, game-UX conventions (menus, pause, onboarding, difficulty, save etiquette), effect-asserting playtests, and the quality bar…
hearth
Operate the Hearth engine as a coding agent — the session loop (recall → snapshot → change → validate → playtest → screenshot → remember), project memory, permission modes, verification, review/undo, and export. Load FIRST whenever a project has a hearth.json or the task mentions Hearth. Routes to the domain skills…
hearth-code
Write behavior in Hearth — Lua/JS scripts, the ctx API (input, timers, tweens, events, camera, audio, save), script modules and require, the dot-call and userdata pitfalls, deterministic RNG, and the check-script/edit-script iteration loop. Use when making things happen in a Hearth game (movement, AI, pickups, rules)…
hearth-design
Shape a complete Hearth game, not just working mechanics — scoping to session length (the verbs × levels × enemies budget), the teach → develop → escalate → climax → end arc, real endings (a win state, not just game-over), one-element-at-a-time difficulty ramps, scenes-as-levels pacing with gated progress and…