tiger-ai-lab/rewardharness/evaluate-physical-plausibility
Claude Code and agent skill by tiger-ai-lab/rewardharness
Trust tier 2
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
Guidelines for evaluating inserted or replaced objects for physical plausibility,
Install
npx vskill@latest install tiger-ai-lab/rewardharness/evaluate-physical-plausibilitySKILL.md
Evaluate Physical Plausibility
When an instruction requires inserting a new object or replacing an existing one (e.g., replacing a remote with a bottle, inserting a person):
- Check for Floating Objects: Ensure the new object is grounded. Objects should not appear to be floating in mid-air above tables, floors, or hands unless explicitly requested.
- Gravity and Interaction: The object must obey basic physics. If it is replacing an object held by a person, it should fit within the grasp or rest naturally on the surface.
- Severe Integration Failures: While minor blending issues are acceptable, a completely floating object or one that defies gravity is a severe failure in physical plausibility and should be heavily penalized, even if the object itself is generated perfectly.
<!-- Release review: RewardHarness v0.4.1 (2026-08-26). -->
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckinglibraries/archive/gemini-v2/skills/evaluate-physical-plausibility/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/tiger-ai-lab/rewardharness/evaluate-physical-plausibility)