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 commands/devdavv/unity-ai-workflow/uw-cmd-polishgit clone --depth 1 https://github.com/devdavv/unity-ai-workflowWhat 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 | $0.00000 | $0.00468 |
| Opus 5 | $0.00000 | $0.00234 |
| Sonnet 5 | $0.00000 | $0.00094 |
| Haiku 4.5 | $0.00000 | $0.00047 |
Grade A, and why
uw-cmd-polish 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 3d 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.
What it actually says
/uw-cmd-polish — Polish Workflow
Apply "juice" to a working feature using the Game Feel Document's Feedback Matrix.
Skills
Load uw-game-feel-integrator from .claude/skills/uw-game-feel-integrator/SKILL.md.
Prerequisites
- The feature must work correctly before polish. Don't polish broken code.
docs/GFD.mdmust have entries for the events you're polishing.- Check
docs/ProjectConfig.yaml -> feel_toolsfor available middleware.
Steps
1. Audit Current Feel
Play the feature and note:
- Which actions feel flat or unresponsive?
- Which moments should be impactful but aren't?
- Is there visual feedback for every player input?
2. Consult the GFD
For each event in the feature, check the Feedback Matrix:
Event: Player Hit
|- VFX: Blood splash, screen flash
|- Audio: Impact SFX (pitch-varied)
|- Camera: Medium shake (0.2s, 0.3 magnitude)
|- Tween: Color flash (red, 0.05s yoyo)
|- Haptic: Strong pulse (0.1s)
When suggesting feel patterns, cite the reference game: "A Celeste-like hang time at jump apex" or "Dead Cells impact freeze."
3. Implement Juice
Use uw-game-feel-integrator skill:
- Screen shake, tweens (scale punch, color flash, position bounce)
- Particle effects, audio cues with pitch variation
- Hitstop (time scale pause) for impactful moments
- Camera zoom for dramatic events
4. Tune Values
- Start exaggerated, then dial back.
- "Too much juice" is easier to fix than "too little."
- Test on target platform — mobile haptics differ from desktop.
5. Performance Check
- Profile the added effects.
- Ensure particle systems are pooled.
- Ensure tweens are properly killed on object destruction.
6. Output
- Polished feature with full feedback loop
- Updated GFD with final tuned values
- Performance validated
$ARGUMENTS
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.
- 3d ago First seen · 57 lines · 0 tokens per session scan A 655cc4101948
uw-cmd-polish is a command published in the GitHub repository devdavv/unity-ai-workflow (41 stars, last pushed 5mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 468 tokens. 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 commands, from other repositories
audio-pass
Functional audio review for implementation, mix coverage, clarity, and player feedback.
godot-review
Review a Godot project for structure, architecture, maintainability, and risk.
level-beat
Design or revise level flow, beats, pacing, and encounter structure.
ui-flow-review
Review menus, HUD, navigation, and player flow from a UX perspective.
start-platform
启动/停止 GameServer 排行榜服务(:22333).
play-dashboard
Play in the dashboard — the local viewer becomes your play surface, with a live AI DM beside it.