← ClaudeAtlas

bld-find-splinelisted

fs: Find a ready-made 3D scene on Spline and wire its embed into the app. Use for "find a 3D scene", "add a Spline", "3D hero". Heavy. Only when it genuinely elevates the page.
mnguyenht/bld-package · ★ 1 · AI & Automation · score 64
Install: claude install-skill mnguyenht/bld-package
# bld-find-spline — shop Spline for a 3D scene, then wire the embed Sibling of `bld-find-21st`, but for **3D scenes on Spline**. Same spirit (find → present → user picks → wire it in), with two big differences baked in below: Spline gates its assets, and 3D is expensive. No MCP server, no API key. Pure WebSearch/WebFetch + npm. ## Hard realities of Spline (read before promising anything) 1. **The community catalog is mostly unreadable.** `community.spline.design` listing/tag pages are login-gated JS shells — WebFetch gets an empty shell. What IS readable: **`spline.design/examples`** (official template gallery, ~18 scenes with `app.spline.design/file/<id>` links) and whatever individual scene URLs a web search surfaces. Discover from those, not the community app. 2. **The user must export the asset themselves.** To actually use a scene, the user logs into Spline, opens/duplicates it, and exports — either a public **scene URL** (`https://prod.spline.design/<id>/scene.splinecode`) or a downloaded **`.splinecode`** file. Free tier = 3 exports/month. You CANNOT do this step; hand it to the user with clear instructions. 3. **3D is heavy.** Scenes are large (often MBs), GPU-hungry, and rough on mobile and battery. This raises the quality bar — see Guardrails. ## Workflow 1. **Pin the need.** What's the 3D for (hero background, product model, icon, playful accent), which app, and the vibe. 3D usually earns its weight only on heroes / landing pa