Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Capture a screenshot from the Unity Editor Scene View at the requested size. Renders via the Scene View's active camera onto a temporary `RenderTexture`. Requires an open Scene View.
.claude/skills/ivanmurzak-screenshot-scene-view/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 94% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -64% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -45% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -43% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -54% | 0% |
Captures a screenshot from the Unity Editor Scene View and returns it as an image. Returns the image directly for visual inspection by the LLM.
width (default 1920) / height (default 1080) — output pixels. Must be > 0 and ≤ MaxDimension.Uses SceneView.lastActiveSceneView (falls back to the first window in SceneView.sceneViews). Allocates a temporary RenderTexture, swaps it onto the Scene View's camera, calls Render, reads back with Texture2D.ReadPixels, encodes PNG, and restores the camera's prior targetTexture. Throws/errors when no Scene View is open or the Scene View camera is null.
bashunity-mcp-cli run-tool screenshot-scene-view --input '{ "width": 0, "height": 0 }'
> For complex input (multi-line strings, code), save the JSON to a file and use: > bash > unity-mcp-cli run-tool screenshot-scene-view --input-file args.json > > > Or pipe via stdin (recommended): > bash > unity-mcp-cli run-tool screenshot-scene-view --input-file - <<'EOF' > {"param": "value"} > EOF >
If unity-mcp-cli is not found, either install it globally (npm install -g unity-mcp-cli) or use npx unity-mcp-cli instead. Read the /unity-initial-setup skill for detailed installation instructions.
| Name | Type | Required | Description | |------|------|----------|-------------| | width | integer | No | Width of the screenshot in pixels. | | height | integer | No | Height of the screenshot in pixels. |
json{ "type": "object", "properties": { "width": { "type": "integer" }, "height": { "type": "integer" } } }
This tool does not return structured output.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 8,224 | 2,319 | -72% | 1 | 1 | 0% | 1,472 | 651 | -56% | 0 | 0 | — |
case-02 | fail→pass | 2,576 | 1,340 | -48% | 1 | 1 | 0% | 372 | 720 | +94% | 0 | 0 | — |
case-03 | fail→fail | 4,668 | 2,335 | -50% | 1 | 1 | 0% | 815 | 707 | -13% | 0 | 0 | — |
case-04 | fail→pass | 14,544 | 2,378 | -84% | 1 | 1 | 0% | 2,863 | 1,019 | -64% | 0 | 0 | — |
case-05 | fail→fail | 15,435 | 7,353 | -52% | 1 | 1 | 0% | 2,741 | 1,782 | -35% | 0 | 0 | — |
case-06 | fail→pass | 8,182 | 1,722 | -79% | 1 | 1 | 0% | 1,477 | 806 | -45% | 0 | 0 | — |
case-07 | pass→pass | 6,945 | 1,483 | -79% | 1 | 1 | 0% | 1,257 | 836 | -33% | 0 | 0 | — |
case-08 | pass→pass | 8,166 | 2,573 | -68% | 1 | 1 | 0% | 1,537 | 1,011 | -34% | 0 | 0 | — |
case-09 | fail→pass | 8,299 | 1,507 | -82% | 1 | 1 | 0% | 1,364 | 779 | -43% | 0 | 0 | — |
case-10 | pass→pass | 10,274 | 2,538 | -75% | 1 | 1 | 0% | 1,920 | 1,000 | -48% | 0 | 0 | — |
case-11 | fail→pass | 13,940 | 5,573 | -60% | 1 | 1 | 0% | 2,461 | 1,138 | -54% | 0 | 0 | — |
case-12 | fail→pass | 14,469 | 2,445 | -83% | 1 | 1 | 0% | 2,574 | 963 | -63% | 0 | 0 | — |
case-13 | fail→pass | 5,457 | 1,683 | -69% | 1 | 1 | 0% | 906 | 785 | -13% | 0 | 0 | — |
case-14 | pass→pass | 11,679 | 1,437 | -88% | 1 | 1 | 0% | 2,131 | 801 | -62% | 0 | 0 | — |
case-15 | fail→pass | 12,293 | 1,673 | -86% | 1 | 1 | 0% | 2,746 | 870 | -68% | 0 | 0 | — |
case-16 | fail→pass | 7,131 | 1,282 | -82% | 1 | 1 | 0% | 1,352 | 727 | -46% | 0 | 0 | — |
case-17 | pass→pass | 11,921 | 4,977 | -58% | 1 | 1 | 0% | 1,892 | 1,407 | -26% | 0 | 0 | — |
case-18 | pass→pass | 19,235 | 6,790 | -65% | 1 | 1 | 0% | 3,176 | 1,802 | -43% | 0 | 0 | — |
case-19 | pass→pass | 14,626 | 13,389 | -8% | 1 | 1 | 0% | 3,015 | 3,136 | +4% | 0 | 0 | — |
case-20 | pass→pass | 15,903 | 9,077 | -43% | 1 | 1 | 0% | 3,173 | 2,383 | -25% | 0 | 0 | — |
case-21 | pass→pass | 11,597 | 6,420 | -45% | 1 | 1 | 0% | 2,375 | 1,833 | -23% | 0 | 0 | — |
case-22 | pass→pass | 9,120 | 6,038 | -34% | 1 | 1 | 0% | 1,919 | 1,775 | -8% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 20 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +41 percentage points is the difference between those two pass rates over the 20 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.