Image to Three.js Model
img2threejs/img2threejs
Rebuilds the object in a reference image as a procedural, animation-ready Three.js model written entirely in code, using staged sculpting with quality checks.
Build guided and interactive virtual tours through a 3D environment with authored camera paths, room and floor-plan navigation, orbit inspection, smooth mode handoffs, and progressive loading.
$ npx skills add MengTo/Skills --skill 3d-virtual-tour -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install MengTo/Skills 3d-virtual-tour --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/MengTo/Skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/agent-skills/3d/3d-virtual-tour .claude/skills/3d-virtual-tour && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "3d-virtual-tour" agent skill from https://github.com/MengTo/Skills/tree/main/agent-skills/3d/3d-virtual-tour into .claude/skills/3d-virtual-tour/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "3d-virtual-tour", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/MengTo/Skills/tree/main/agent-skills/3d/3d-virtual-tourType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add MengTo/Skills --skill 3d-virtual-tour -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install MengTo/Skills 3d-virtual-tour --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/MengTo/Skills.git skills-src && mkdir -p .agents/skills && cp -r skills-src/agent-skills/3d/3d-virtual-tour .agents/skills/3d-virtual-tour && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "3d-virtual-tour" agent skill from https://github.com/MengTo/Skills/tree/main/agent-skills/3d/3d-virtual-tour into .agents/skills/3d-virtual-tour/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "3d-virtual-tour", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add MengTo/Skills --skill 3d-virtual-tour -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install MengTo/Skills 3d-virtual-tour --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/MengTo/Skills.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/agent-skills/3d/3d-virtual-tour .cursor/skills/3d-virtual-tour && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "3d-virtual-tour" agent skill from https://github.com/MengTo/Skills/tree/main/agent-skills/3d/3d-virtual-tour into .cursor/skills/3d-virtual-tour/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "3d-virtual-tour", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/MengTo/Skills.git --path agent-skills/3d/3d-virtual-tour--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add MengTo/Skills --skill 3d-virtual-tour -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install MengTo/Skills 3d-virtual-tour --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/MengTo/Skills.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/agent-skills/3d/3d-virtual-tour .gemini/skills/3d-virtual-tour && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "3d-virtual-tour" agent skill from https://github.com/MengTo/Skills/tree/main/agent-skills/3d/3d-virtual-tour into .gemini/skills/3d-virtual-tour/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "3d-virtual-tour", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install MengTo/Skills 3d-virtual-tourInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add MengTo/Skills --skill 3d-virtual-tour -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/MengTo/Skills.git skills-src && mkdir -p .github/skills && cp -r skills-src/agent-skills/3d/3d-virtual-tour .github/skills/3d-virtual-tour && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "3d-virtual-tour" agent skill from https://github.com/MengTo/Skills/tree/main/agent-skills/3d/3d-virtual-tour into .github/skills/3d-virtual-tour/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "3d-virtual-tour", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add MengTo/Skills --skill 3d-virtual-tour -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install MengTo/Skills 3d-virtual-tour --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/MengTo/Skills.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/agent-skills/3d/3d-virtual-tour .opencode/skills/3d-virtual-tour && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "3d-virtual-tour" agent skill from https://github.com/MengTo/Skills/tree/main/agent-skills/3d/3d-virtual-tour into .opencode/skills/3d-virtual-tour/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "3d-virtual-tour", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
3d-virtual-tourBuild guided and interactive virtual tours through a 3D environment with authored camera paths, room and floor-plan navigation, orbit inspection, smooth mode handoffs, and progressive loading.
3D Virtual Tour is an agent skill from MengTo/Skills. Build guided and interactive virtual tours through a 3D environment with authored camera paths, room and floor-plan navigation, orbit inspection, smooth mode handoffs, and progressive loading. Use for architectural walkthroughs, real-estate tours, museums, showrooms, and explorable spaces built with Three.js or a similar renderer.
Its SKILL.md is about 2.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files (for example `REFERENCES.md` and `agents/openai.yaml`).
It sits in Business, Finance & HR, covering Real estate and 3D graphics and WebGL. It works with Three.js. The repository describes itself as: Agent skills for designers and builders using Codex, Claude, Cursor, and other AI coding agents. The licence is MIT.
Read from SKILL.md and the folder at commit 83a47fe. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
Links to these hosts (documentation or services it may open):
threejs.orgFrom URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
3D Virtual Tour loads about 2.3k tokens when it runs. Until then it costs about 87 tokens; SKILL.md has 1,241 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from MengTo/Skills at commit 83a47fe, republished under its MIT licence (© MengTo). 1,241 words, ~2,275 tokens.
.claude/skills/3d-virtual-tour/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.Help the viewer understand a place through movement, landmarks, and direct navigation. Keep the route, camera, room labels, and controls in agreement.
Inspect the scene, coordinate system, scale, floor elevations, door openings, renderer, camera controls, and asset-loading path. Preserve existing architecture and interactions. Choose the navigation modes the experience actually needs:
| Mode | Camera owner | Navigation |
|---|---|---|
| Guided tour | Authored route sampler | Scroll, play/pause, or previous/next stops |
| Room destination | Authored room pose or route stop | Room list, hotspot, or floor plan |
| Inspect | Bounded orbit controller | Drag, pinch, zoom, and return |
| Free walk, when requested | First-person controller with collision | Movement and look controls |
For a Seijaku-style experience, start with a guided route, room destinations, and optional model inspection. Add free walking only when requested or required; orbit controls alone do not provide a walkable interior. If the source is only a set of panoramic images, use linked panorama viewpoints rather than implying that it contains navigable geometric depth.
Store route keys as progress/time, position, look target or orientation, and field of view. Store stops with stable IDs, labels, route coordinates, and associated asset groups. Keep the route coordinate independent of page height or animation clock so room links survive layout changes.
Place keys before, within, and after narrow openings. Use floor-relative eye height, account for raised floors and steps, and slow before turns or important views. Keep the horizon stable and avoid large simultaneous changes in position, gaze, and field of view. Favor physically plausible proportions over an excessively wide lens that stretches rooms.
Use timed Hermite interpolation, a suitable spline, or authored animation. Seijaku uses nonuniform timed Hermite keys for position, gaze, and field of view. A centripetal spline is another useful option, but any smooth curve can cut a corner between collision-free keys. Inspect the complete interpolated path, including the camera's near-plane clearance, against walls, ceilings, furniture, and doorframes.
Curve parameter is not automatically distance or travel time. Use arc-length sampling when roughly constant speed matters, or intentionally time segments to create pauses and reveal landmarks. Inspect Curve for the installed renderer's parameterization methods.
Interpolate orientation with quaternion slerp when using orientation keys. If interpolating look targets, prevent the target from crossing or coinciding with the camera, which can flip its direction. Update the projection matrix when field of view or aspect changes.
Use a small navigation state machine and one final camera write per frame. Inputs request state changes; the route sampler, orbit controller, and transition handler must not all mutate the live camera independently.
Entering inspection should save the current route coordinate, active stop, playback state, and any scroll bookmark. Initialize inspection from a compatible view on the viewer's current side of the building. Freeze guided progression while inspecting.
Blend position, orientation, and field of view during the handoff using a validated transition path. A straight line from an interior to an exterior orbit can cross a roof; even an upward arc needs checking. Use safe waypoints, or a brief fade and destination cut when there is no credible continuous route. Clamp orbit distance and pitch to the intended inspection area.
Returning restores the saved guided state, reconciles scroll position, clears stale drag velocity, and resumes playback only if it was previously running. Disable inactive controllers and stale damping updates so they cannot pull the camera away after a transition. New navigation requests should replace the pending destination from the current pose rather than queue several competing tweens.
Drive room cards, a chapter rail, floor-plan regions, and scene hotspots from the same stop registry. Use a clear world-to-plan coordinate transform; do not assume a plan's vertical axis matches world Z. Highlight the current room or route segment, with hysteresis at boundaries to avoid flicker.
Room selection should either travel along a safe connecting route or use an intentional cut/fade. Do not fly a direct interpolation through several walls. Keep the selected label and URL state, if present, synchronized with the actual destination; update browser history on meaningful navigation rather than every frame.
Use semantic HTML buttons or links for stops and provide a keyboard-accessible room list alongside graphical maps. Label hotspots, keep focus visible, and restore focus to the inspection trigger on exit. Make Escape and a visible Return button release inspection mode.
On a scrolling page, touch gestures should continue scrolling until the viewer explicitly enters interaction mode. Scope pointer capture and touch-action to the active canvas region. Handle pointer cancellation, lost capture, and blur; restore the prior scrolling styles on exit. Do not intercept wheel or keyboard input over forms or unrelated page controls.
Derive route-driven doors, captions, and exposure from the current route coordinate so seeking backward or jumping to a room produces the correct scene immediately. Timer-only events can leave a closed door in the camera's path after a seek. Preserve user-selected seasons and lighting overrides unless the tour explicitly owns those settings.
Prefetch upcoming room assets by route proximity and load a directly selected destination before revealing it. Show a useful placeholder or hold a stable view during loading; a failed optional asset should not trap navigation. Cache shared textures and meshes, and cancel or ignore obsolete destination work after a new request.
Use region visibility only when it remains correct for the current camera mode. Regions hidden along a guided path may become visible when the viewer orbits the model. Recompute visibility or enable the required regions during inspection instead of exposing missing walls or rooms.
If free walk is included, constrain movement to walkable surfaces and provide collision/clearance checks. Pointer lock supplies look input, not collision or floor following. Give touch users an equivalent navigation path and a visible exit from immersive controls.
Use frame-time-based interpolation and reset the time base after hiding the page. Pause playback and unnecessary rendering while hidden. Retain a static, informative view under reduced motion; room buttons should select composed views without requiring a long camera flight. Start audio only after a user gesture and preserve its mute state across navigation.
Keep one render loop, reuse scratch objects, and avoid layout reads or large allocations inside pointer handlers. Prepare the opening room first and spread expensive scene generation, texture uploads, and shader preparation across appropriate stages. Resize camera framing and render targets together; tune narrow-view compositions without moving the camera into geometry. Remove listeners, release capture, and dispose owned resources when the tour unmounts.
Read REFERENCES.md for Seijaku's KEYS, interp, Film, Orbit, _rooms, _plan, and goTo implementation, plus the relevant camera APIs. Its guided walk and exterior inspection are reference techniques. Collision-based free walking and the additional accessibility checks above are implementation guidance, not claims about the reference's features.
© MengTo, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 2 other files in agent-skills/3d/3d-virtual-tour of MengTo/Skills.
Open the folder on GitHubat commit 83a47fe
3D Virtual Tour next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| 3D Virtual Tour this skillMengTo/Skills | 6.7k | — | ~2.3k | Automated safety check: Pass | MIT | |
| Image to Three.js Modelimg2threejs/img2threejs | 18k | 1 repos | ~8.2k | Automated safety check: Pass | Apache-2.0 | |
| Web CloneJane-xiaoer/claude-skill-web-clone | 1k | 1 repos | ~2.7k | Automated safety check: Pass | MIT | |
| Threejs Game Directormajidmanzarpour/threejs-game-skills | 2.5k | — | ~2.2k | Automated safety check: Pass | MIT | |
| Threejs Gameplay Systemsvalkor-ai/loom | 1.2k | 1 repos | ~1.4k | Automated safety check: Pass | Apache-2.0 | |
| Threejs World Generationcalesthio/OpenMontage | 66k | — | ~2k | Automated safety check: Pass | AGPL-3.0 |
img2threejs/img2threejs
Rebuilds the object in a reference image as a procedural, animation-ready Three.js model written entirely in code, using staged sculpting with quality checks.
Jane-xiaoer/claude-skill-web-clone
网站复刻 / 克隆方法论。USE WHEN 用户说 复刻网站、克隆网站、clone website、抄个站、仿站、 照着这个站做一个、reproduce site、还原某个网页效果、把这个站搬下来改成我的、 复刻某个交互/WebGL/Canvas/Three.js 效果。提供「先拿真源码 → 判路径 → 逆向拆解 → 搭工程 → 替换内容」的可移植决策树,覆盖静态站 /…
majidmanzarpour/threejs-game-skills
Entrypoint for building, upgrading, and finishing Three.js browser games.
valkor-ai/loom
Build and iterate playable Three.js game systems: starter scaffold, architecture, design briefs, core loops, level and encounter design, entities, input, camera, collision and physics, scoring…
calesthio/OpenMontage
Build deterministic, editable, free-viewpoint Three.js worlds from text or structured briefs.
wy51ai/edulab
把一道解析几何题解成一个自包含的交互教学网页:左栏题面 + 动态控制台(一个 可变参数滑块驱动实时重算的几何量 + 理论范围/定值指示),中栏 KaTeX 分步解析,右栏 2D Canvas 动态几何画板(椭圆/双曲线/抛物线/圆 + 动直线/动点 + 向量 + 标注 + 画笔涂鸦)。
MengTo/Skills
Draws hand-authored SVG spot illustrations in a bold pop-geometric style for design and coding scenes.
MengTo/Skills
Draws a slim, long-limbed character acting out one coding or design moment as an outline-free flat SVG card, in a fixed measured palette with confetti and no ground line.
MengTo/Skills
Draws flat vector spot illustrations of lanky people doing design and coding things, with solid black as a main fill, for empty states, onboarding and error pages.
MengTo/Skills
Draws cozy animal scenes as hand-authored 480x360 SVG cards in a grainy, risograph-like gouache style on hot pink, for empty states, headers and marketing cards.
MengTo/Skills
Draws hand-authored SVG spot illustrations in a halftone line style for design and coding scenes.
MengTo/Skills
Draws hand-inked cartoon critters and living objects as SVG cards (480x360), each acting out a coding or design idea.
Works with
Categories
Build guided and interactive virtual tours through a 3D environment with authored camera paths, room and floor-plan navigation, orbit inspection, smooth mode handoffs, and progressive loading. 3D Virtual Tour is an agent skill from MengTo/Skills. Build guided and interactive virtual tours through a 3D environment with authored camera paths, room and floor-plan navigation, orbit inspection, smooth mode handoffs, and progressive loading.
3D Virtual Tour fits situations like: architectural walkthroughs; real-estate tours; explorable spaces built with Three.js; A similar renderer.
Run `npx skills add MengTo/Skills --skill 3d-virtual-tour -a claude-code`. Or copy the skill folder (agent-skills/3d/3d-virtual-tour in MengTo/Skills) into .claude/skills/3d-virtual-tour in your project. Claude Code loads it when a task matches its description.
Run `npx skills add MengTo/Skills --skill 3d-virtual-tour -a codex`. Or copy the skill folder (agent-skills/3d/3d-virtual-tour in MengTo/Skills) into .agents/skills/3d-virtual-tour in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add MengTo/Skills --skill 3d-virtual-tour -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/3d-virtual-tour, .gemini/skills/3d-virtual-tour, .github/skills/3d-virtual-tour and .opencode/skills/3d-virtual-tour in your project.
SKILL.md names no scripts, command-line tools or credentials: 3D Virtual Tour is instructions for the agent only.
SKILL.md names 1 domain. As links in the text: threejs.org. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
3D Virtual Tour is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 2.3k tokens (SKILL.md is roughly 9.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with 3D Virtual Tour: Image to Three.js Model (img2threejs/img2threejs, 18k stars), Web Clone (Jane-xiaoer/claude-skill-web-clone, 1k stars), Threejs Game Director (majidmanzarpour/threejs-game-skills, 2.5k stars) and Threejs Gameplay Systems (valkor-ai/loom, 1.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
MengTo (a GitHub user) maintains it in MengTo/Skills, which has 6,675 GitHub stars. The repository holds 39 skills in this directory. The repository was last updated on October 6, 2026.
Source: MengTo/Skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.