Borrowing it
Nothing to install: this file belongs to ukanwat/aaabench. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/ukanwat/aaabench/main/.claude/skills/physics-tuning/SKILL.mdgit clone --depth 1 https://github.com/ukanwat/aaabenchWrote 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/ukanwat/aaabench/physics-tuning)<a href="https://agentmods.dev/skills/ukanwat/aaabench/physics-tuning"><img src="https://agentmods.dev/badge/skills/ukanwat/aaabench/physics-tuning/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/ukanwat/aaabench/physics-tuning"><img src="https://agentmods.dev/badge/skills/ukanwat/aaabench/physics-tuning.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 | $0.00083 | $0.01841 |
| Opus 5.5 | $0.00033 | $0.00736 |
| Sonnet 5.5 | $0.00017 | $0.00368 |
| Haiku 4.5 | $0.00008 | $0.00184 |
Grade A, and why
physics-tuning 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.
This is a copy
89% identical to physics-tuning — 37 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Physics tuning
Most "bad physics" is not a bug in the engine — it's a mismatch between the
fixed-timestep simulation and the variable-rate render loop, or untuned
mass/drag/CCD/layer settings. This skill covers the engine-neutral knobs that
make physics stable and responsive; pair it with godot-physics or
unity-physics for the concrete APIs.
When to use
- Use when motion jitters, objects pass through walls (tunneling), stacks explode, or movement feels floaty/sticky/laggy.
- Use to decide what goes in the fixed (physics) step vs the render frame, and how to interpolate between them.
- Use to tune gravity, mass, drag, restitution, solver iterations, sleeping, and collision layers/masks.
When not to use: for an engine's exact physics nodes/components and
collision callbacks, use godot-physics or unity-physics. For movement
decisions (when to jump, AI steering) use input-systems and game-ai. For
platformer jump-feel specifics like coyote time/jump buffering, that's input/
controller territory — see input-systems and the platformer genre.
Core workflow
- Run physics on a fixed timestep. Simulate at a constant rate (e.g. 50–60
Hz). A fixed
dtmakes the simulation deterministic-ish and stable; a variabledtmakes integration and collisions inconsistent. - Put physics work in the physics callback, not the render frame. Apply
forces/velocities and read collisions in the fixed step (
FixedUpdate/_physics_process), using that step'sdt. - Interpolate rendering between physics ticks. The render frame rate ≠ the physics rate, so smoothly interpolate transforms toward the latest physics state, or enable the engine's Rigidbody interpolation, to remove visible stutter.
- Tune the body, not the scene. Set mass for relative weight, drag for damping, gravity scale per object, and restitution/friction via materials.
- Stop tunneling with CCD on small/fast bodies; cap maximum velocity.
- Stabilize stacks/joints with more solver iterations, sane mass ratios, and sleeping for resting bodies.
- Verify by feel and stress test. Play at low and high frame rates; throw fast objects at thin walls; stack and shove bodies. Report what you observed.
What ships with it
1 file 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.
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 · 156 lines · 83 tokens per session scan A b1372a28ee6d
physics-tuning is a skill published in the GitHub repository ukanwat/aaabench (385 stars, last pushed 2mo ago), licensed MIT. It adds 83 tokens to every session and 1,841 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 89% identical to physics-tuning, differing in 37 lines, and is treated as a copy.
Other skills, from other repositories
unreal-engineer
!cat skills/shared/protocols/3d-spatial-foundations.md 2>/dev/null || echo "=== 3D Foundations not loaded ===".
unreal-playtest-agent
Domain skill - run screenshot-light PIE playtest episodes with structured entity observations, bounded semantic actions, transition polling, and in-memory traces for QA and external policy or RL runners.
unreal-level
Domain skill - Unreal Engine level and world management: inspect level state, load and save maps, read or edit world settings, and measure native collision distances in Editor or PIE worlds. Not for individual actor transform edits - use unreal-actors for that.
unreal-pie
Domain skill - Unreal Engine Play-In-Editor (PIE) closed-loop verification. Provides PIE lifecycle control (enter/pause/resume/exit), controlled input injection (no OS dependency), viewport screenshot capture, output log snapshot, performance sampling, and Automation Test execution with async job polling. Use for…
unreal-actors
Domain skill - Unreal Engine actor management: list, spawn, transform, and delete level actors. Use when inspecting or editing actors in the current Unreal Editor level. Not for Content Browser asset import/export - use unreal-assets for that.
unreal-assets
Domain skill - Unreal Engine Content Browser asset management: list, import, export, inspect, and delete assets. Use for package and asset operations in the Unreal Editor. Not for actor placement or transforms - use unreal-actors for that.