| FMRS | 13 / 100 · D | 79 / 100 · B |
|---|
| Reliability | 2 / 20 | 12 / 20 |
|---|
| Security and permissions | 2 / 20 | 18 / 20 |
|---|
| Maintenance | 3 / 20 | 16 / 20 |
|---|
| Documentation | 2 / 20 | 18 / 20 |
|---|
| Setup experience | 4 / 20 | 15 / 20 |
|---|
| Best for | - Developers building SceneView-based 3D/AR interior-design features
- Teams that want AI-assisted generation of room planners, furniture configurators, and lighting previews
- MCP workflows already using Claude Code, Claude Desktop, Cursor, or Windsurf
| - Engineering teams scaling AI coding agents while controlling spend
- Platform teams that need auditable, rollback-aware agent execution records
- Organizations using multiple agents (Claude, Codex, Gemini) under one governed flow
|
|---|
| Not for | - General-purpose 3D or AR development — use the main sceneview MCP server (28 tools) instead
- Other verticals such as automotive, healthcare, gaming, or Rerun.io debugging (separate specialty servers)
- Projects that need exact, pre-documented tool names — the sources do not list this server's 7 tool names
- Scenarios where other 5 MCP servers in the same repo would be confused with this one
| - Developers who just want a simple code-completion helper without run governance
- Teams unwilling to accept local JSONL run records with HMAC signing
- Environments without a supported adapter (Claude/Codex/Gemini CLIs, direct provider, or verifier-only)
|
|---|
| Required permissions | - Runs as a local child process via npx, so it inherits the permissions of the launching client or shell (the README describes no restricted permission model)
- May need read/write access to files in the current project if generate/analyze-style tools are invoked
- Network access is required to download the npm package and potentially SceneView assets
| - Read/write access to the configured repo root (writable scope restrictable via --allow-path/--deny-path)
- Ability to invoke the chosen agent CLI (Claude/Codex/Gemini, etc.) and its underlying model account
- Local disk writes for run records, receipts, and share outputs (share/run-receipt., etc.)
- Execution of user-configured verifier commands (e.g. npm test)
|
|---|
| Risks and side effects | - Running `npx` executes third-party code on your machine; pin versions and review the package before use
- This server is maintained as part of the SceneView repository and is not an official Anthropic MCP server; verify its maintenance status
- AI-generated SceneView code may require a matching SDK version (e.g. 4.26.0) and platform-specific setup (Android/Filament, iOS/RealityKit, Web) to compile
- Because tool names are undocumented in the supplied material, tool calls may fail if the client expects specific identifiers
- Easy to confuse with the other 5 MCP servers in the repo — this profile covers only interior-design-3d
| - Verifier commands are executed; poorly chosen verifier commands carry security risk despite preflight policy checks
- Cost and token figures may be estimates (labeled with provenance) and should not be treated as settled accounting
- Receipt integrity must be verified before use as trustworthy evidence; missing rollback evidence is flagged as EVIDENCE_BOUNDARY
- Policy defaults come from martin.config.yaml and can be overridden by CLI flags; misconfiguration can loosen safety boundaries
|
|---|
| Supported clients | Claude Desktop, Claude Code, Cursor, Windsurf | Claude Code, Codex, Gemini CLI, Cursor, VS Code, GitHub Copilot, Continue |
|---|
| Tools | 0 | 0 |
|---|