Sceneview MCP

Official

Official MCP server for the SceneView 3D/AR SDK, offering API docs, samples, validation, and code generation.

How to install ↓ Compare GitHub source repository ↗

Category: Dev Tools #327 of 438
Stars: ★ 1.3k Very popular
Transport: stdio (local process)
Runtime: Node.js
Credentials: Optional API key
License: Apache-2.0
Last commit: 1 day ago
Tools: 5

43FMRS · D

This MCP server is officially maintained by SceneView and focuses on AI-assisted 3D/AR code generation and validation. It provides rich tools and documentation, making it highly valuable for SceneView users. However, its scope is SDK-specific, which may not suit general-purpose needs.

Strongest · Setup experience 12/20 Weakest · Reliability 6/20

Reliability

6/20

Security and permissions

6/20

Maintenance

10/20

Documentation

9/20

Setup experience

12/20

Why each score

Reliability 6/20: Only the manifest and README were supplied; no actual mcp/server.json implementation, tool list, tests, or CI configuration was provided, so a clean startup, init handshake, and tool behavior cannot be verified. README claims of 28 tools and 33 samples are untrusted evidence and the score is conservatively lowered.
Security and permissions 6/20: No red-line issues (malware, embedded secrets, destructive defaults) are visible in the supplied material; however, the README mentions a BYOK Sketchfab search_models tool, and API-key handling, least-privilege boundaries, external data flows, and confirmation of risky operations are entirely undocumented, so the score is deducted for missing evidence.
Maintenance 10/20: The repository is not archived, the Apache-2.0 license and npm package version 4.0.12 appear in the manifest, and the README shows CI, a changelog, and community channels; but no commit history, issue-response records, or dependency-update evidence was supplied, so the score is moderate rather than high.
Documentation 9/20: The README provides a layered SDK overview, platform tables, installation options, and names several MCP tools such as generate_scene and validate_code; however, complete tool parameter schemas, authentication details, rate/cost limits, and troubleshooting guidance for the MCP server are missing, and the referenced mcp/README.md was not provided.
Setup experience 12/20: The install path is clear and few-step: npx sceneview-mcp over stdio, with examples for Claude Code and JSON MCP configs in mainstream clients. However, the static calibration requires execution-level evidence (CI plus tests) for a higher setup score, and Node version/platform requirements are unspecified, so the score is capped conservatively.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

Developers using the SceneView SDK
Developers who want to generate correct and compilable 3D/AR code in AI tools like Claude, Cursor
Teams that need quick prototyping with samples

Not for

Users who do not use the SceneView SDK
Those needing a general-purpose MCP server beyond graphics rendering
Developers using other 3D SDKs (e.g., Unity, Unreal)

Required permissions

No special system permissions
May prompt the user for a Sketchfab API key (optional)
Access to project files (via `analyze_project`) requires user-granted file access

Risks and side effects

Improper tool usage may generate incorrect code, but `validate_code` mitigates this
Searching models may require an external API key; handle keys securely
Generated code may depend on specific SDK versions; verify compatibility

Setup

Before you start

Runtime：Node.js

For Claude Code, run: claude mcp add sceneview -- npx sceneview-mcp. For Claude Desktop / Cursor / Windsurf, add the following to your MCP config: {"mcpServers":{"sceneview":{"command":"npx","args":["-y","sceneview-mcp"]}}}.

claude_desktop_config.json

{"mcpServers":{"sceneview":{"command":"npx","args":["-y","sceneview-mcp"]}}}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json

{
  "servers": {
    "sceneview": {
      "command": "npx",
      "args": [
        "-y",
        "sceneview-mcp"
      ]
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal

claude mcp add sceneview -- npx -y sceneview-mcp

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

Add {"mcpServers":{"sceneview":{"command":"npx","args":["-y","sceneview-mcp"]}}} to your client's MCP config; the tool list should show generate_scene, debug_issue, search_models, analyze_project and validate_code, or ask "Build me an AR app with SceneView" and receiving compilable Compose code confirms the connection. Note search_models uses Sketchfab BYOK, so a Sketchfab API key is needed to use that tool.

Troubleshooting

Ensure Node.js is installed and npx is available
Check network connectivity
If `search_models` fails, verify the Sketchfab API key
Verify MCP configuration is correct, including paths and arguments
Check MCP server logs to diagnose startup issues

Things to try

Once connected, you can ask your AI assistant things like:

Build me an AR app with tap-to-place furniture using SceneView
Validate this SceneView Kotlin code before I compile it
Search for a helmet.glb 3D model I can use in a scene
Analyze my existing Android project and audit its 3D/AR integration

Tools 5

generate_scene writes

Generate code for a 3D or AR scene based on a description.

debug_issue read-only

Analyze errors or issues and provide debugging suggestions.

search_models read-only

Search for 3D models on Sketchfab (requires API key).

analyze_project read-only

Audit an existing app and provide improvement suggestions.

validate_code read-only

Compile-check generated code before sending it.

Use cases

Generate 3D/AR code to accelerate prototyping

Debug existing SceneView projects

Search for 3D model assets and integrate them

Audit existing apps to optimize performance

Validate code before submission to ensure compilability

Supported clients

Claude Desktop

Claude Code

Cursor

Windsurf

Listed from the project's documentation, not tested by this site.

Overview

This MCP server provides developer tools for the SceneView SDK, covering Android, iOS, and Web platforms. It offers 28 tools, 33 compilable samples, a full API reference, and a code validator to help AI assistants generate correct, compilable 3D/AR code on the first try. Highlights include generate_scene, debug_issue, search_models, analyze_project, and validate_code. It runs on stdio via the npm package sceneview-mcp.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →

NPM Sentinel MCP Server 79 · B

AI-powered NPM package analysis MCP server

★ 18 · Tools 19 Compare with this →

MartinLoop MCP Server 79 · B

An independent control layer that gives AI coding agents budgets, brakes, and receipts

★ 410 Compare with this →

GitHub MCP Server 76 · B

GitHub's official MCP server for managing repos, issues, PRs, and workflows via natural language

★ 33.5k · Tools 15 Compare with this →

Compare with other MCP servers

More servers from this repository

Source revision 5867681d6163 Data synced 2026-10-12 Read the FMRS scoring method