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/hashgraph-online/awesome-codex-plugins/unity-agent-workflowsnpx skills add hashgraph-online/awesome-codex-plugins --skill unity-agent-workflowsgit clone --depth 1 https://github.com/hashgraph-online/awesome-codex-pluginsWhat 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 | $0.00146 | $0.04017 |
| Opus 5 | $0.00073 | $0.02008 |
| Sonnet 5 | $0.00029 | $0.00803 |
| Haiku 4.5 | $0.00015 | $0.00402 |
Grade A, and why
unity-agent-workflows 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 — 242 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Unity Agent Workflows
AI contract for Unity work. Keep answers compact, but never remove exact paths, object/class names, commands, errors, validation, rollback notes, or proof requirements.
Core Loop
- No proof, no edit.
- Treat this injected
SKILL.mdas authoritative; do not search for its own path unless missing or asked for diagnostics. - Read live project instructions narrowly:
AGENTS.mdif present,git status --short, and existingUNITY_STRUCTURE.mdplus only the focused map that matches the task when available. ReadREADME.md, architecture docs,.agents/*, or repo-local agent files only when the task cites them, needs architecture/new-file/asmdef/routing decisions, or maps are missing/stale. - Run
git status --short; preserve unrelated dirty files. - If user says only
$unity-agent-workflows. Teach, run Teach. - Derive only relevant live structure before architecture claims: folders, namespaces,
.asmdef, scenes/prefabs, bootstraps, graphs, docs,UNITY_STRUCTURE*. - Classify task with
Task Gate, then load required references before edits. - Inspect real owner files; never start from memory, nearest-name guessing, semantic similarity, or screenshot proximity alone.
- Fill useful
Routing Cardfields for structure, visible output, state flow, multi-agent, or hub changes. - Patch the smallest proven owner/artifact.
- Validate, sync mirrors if package skill files changed, then close out with proof.
Task Gate
| Trigger | Required action |
|---|---|
| Runtime/visible bug | Prove owner chain before editing. |
| Runtime-visible output, target alignment, focus/highlight/marker/HUD, overlay, input blocker, modal dimming, duplicate names, hardcoded layout, "do not guess" | Runtime Visible Output Hard Stop. |
| Screenshot or visible UI text fix | Screenshot Text Owner Gate; search exact visible text/localization key and prove creator plus refresh writer before editing. |
| Visual/model/sprite/asset integration where feature name, source asset, factory, or visible surface may disagree | Visible Object Identity Lock, Multi-Surface Visible Behavior Lock, and Single-Agent Anti-Anchoring Guard before editing. |
| Same visible behavior requested across multiple scenes/surfaces, or words like "also", "after transition", "menu and gameplay", "preview and runtime", "scene A and scene B" | Multi-Surface Visible Behavior Lock; prove each surface owner, active caller, asset/factory dependency, and fallback behavior before editing any surface. |
| Shared factory/helper/style/global method candidate for a scoped visible surface | Shared Caller Blast Radius Gate: prove allowed runtime caller(s), all other callers, and whether the patch is surface-local before editing. |
| Overlay/dim source-bound mistakes | Prove source target bounds; overlay/dim/mask/blocker rects are destination output unless explicit marker proof exists. |
| Repeated visible-output mismatch after patch | Runtime numeric proof before another coordinate/layout/fallback patch. |
| Multi-agent visible-output or state work | Sub-agent decision and permission gate first; main-agent scope lock before workers patch; checker must review. |
| New/moved files, expanded C# responsibility, asmdef/module routing, hub deflation | Project-derived routing, no fixed structure, hub stop gate. |
| UI, screenshot, HUD, menu, safe area, TMP, spotlight, item-based UI assembly, picked UI parts, UI slices, extracted UI parts, textless UI sprites, design manifests | UI workflow plus runtime-owner proof, extracted item assembly proof, and validation. |
| Visual source asset | Source asset/generator gate before Unity integration code. |
| Content, progression, economy, levels, objectives, gameplay tuning | Data-first content workflow. |
| Tutorial, onboarding, objective, unlock, selection, action, navigation, state transition, guided selection/action flows | State-step guards. |
| Compile/runtime doubt, stale Bee/Roslyn response files, Play Mode proof | Validation workflow. |
| Cleanup/deletion/generated files/git hygiene | Cleanup proof. |
| Rule/session mining/workflow update | Patch only the owning artifact. |
What ships with it
31 files 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.
- .agents/plugins/marketplace.json 402 B
- .codex-plugin/plugin.json 2.0 KB
- .codexignore 288 B
- assets/unity-workflows.png 2.0 KB
- LICENSE 1.0 KB
- package.json 1.5 KB
- README.md 19 KB
- SECURITY.md 1007 B
- skills/unity-agent-workflows/agents/openai.yaml 394 B
- skills/unity-agent-workflows/references/ai-workflows.md 6.6 KB
- skills/unity-agent-workflows/references/asset-source-lock.md 1.3 KB
- skills/unity-agent-workflows/references/cleanup-and-git.md 1.7 KB
- skills/unity-agent-workflows/references/content-and-systems.md 4.3 KB
- skills/unity-agent-workflows/references/coordinate-space-conversion.md 6.6 KB
- skills/unity-agent-workflows/references/modular-architecture.md 5.0 KB
- skills/unity-agent-workflows/references/multi-surface-visible.md 1.0 KB
- skills/unity-agent-workflows/references/project-structure-discovery.md 5.7 KB
- skills/unity-agent-workflows/references/runtime-numeric-proof.md 2.2 KB
- skills/unity-agent-workflows/references/runtime-owner-proof.md 4.7 KB
- skills/unity-agent-workflows/references/runtime-visible-output.md 1.8 KB
- skills/unity-agent-workflows/references/runtime-visible-targets.md 8.1 KB
- skills/unity-agent-workflows/references/screenshot-text-owner.md 1.1 KB
- skills/unity-agent-workflows/references/serialized-persistence.md 1.7 KB
- skills/unity-agent-workflows/references/session-mining.md 2.1 KB
- skills/unity-agent-workflows/references/shared-caller-blast-radius.md 2.0 KB
- skills/unity-agent-workflows/references/target-bounds-catalog.md 3.3 KB
- skills/unity-agent-workflows/references/ui-and-visual-assets.md 9.2 KB
- skills/unity-agent-workflows/references/unity-validation.md 6.7 KB
- skills/unity-agent-workflows/references/visible-object-identity.md 1.8 KB
- skills/unity-agent-workflows/references/workflow-recipes.md 7.0 KB
- skills/unity-agent-workflows/SKILL.md 19 KB
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 · 242 lines · 146 tokens per session scan A eb15bd056553
unity-agent-workflows is a skill published in the GitHub repository hashgraph-online/awesome-codex-plugins (859 stars, last pushed 3d ago), licensed Apache-2.0. It adds 146 tokens to every session and 4,017 once invoked, about $0.0007 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 2026-08-30.
Other skills, from other repositories
search
Search 2500+ curated ChatGPT and LLM open-source repositories. Use when the user asks to find tools, libraries, or repos related to ChatGPT, LLMs, RAG, agents, langchain, NLP, AI development, or any open-source AI tooling.
performance-optimization
Find and fix game performance problems methodically — measure with the engine profiler first, reason about the frame-time budget, locate the CPU-vs-GPU bottleneck, then apply the right fix: object pooling, draw-call batching, fewer allocations/GC spikes, and asset budgets. Engine- neutral method that pairs with each…
ai-behavior-trees-utility-ai
Build a production behavior-tree runtime (Blackboard, action/condition leaves, sequence/selector/parallel composites, decorators) and a Utility AI system (response curves — linear, exponential, sigmoid, quadratic — considerations, and action evaluators), plus hybrid BT-drives-Utility agents. Use when implementing a…
audio-design
Implement game audio practice — bus/mixer architecture and gain in decibels, ducking (sidechain), adaptive/dynamic music via layering and re-sequencing, SFX variation, and beat synchronization. Engine-neutral. Use when the user mentions audio mixing, audio buses, adaptive/dynamic music, ducking, SFX variation, music…
create-game-assets
Plan, generate, source, normalize, and validate cohesive visual game assets. Use for art direction, style bibles, sprites, tilesets, backgrounds, UI art, icons, textures, concept art, or 3D asset briefs.
input-systems
Architect game input — action mapping (abstracting keys into named actions), rebinding with conflict detection and persistence, multi-device support (keyboard, gamepad, touch), analog deadzones, and feel features like input buffering and coyote time, plus accessibility. Engine-neutral. Use when the user mentions input…