M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
R

Rerun 3D

Community
MCP server for debugging SceneView AR sessions with Rerun.io spatial visualization.
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #405 of 438
Stars
★ 1.3k Very popular
Transport
stdio (local process)
Runtime
Node.js
Credentials
No credential needed
License
Apache-2.0
Last commit
32FMRS · D

This is a small, stdio-based MCP server shipped inside the SceneView repository to bridge SceneView AR apps with Rerun.io. The manifest names it io.github.sceneview/rerun-3d (npm package rerun-3d-mcp v1.0.0). The README attributes five tools to the Rerun.io integration, focused on AR debug logging, Python sidecars, web viewer embeds, and spatial data visualization. Tool identifiers are not documented in the supplied sources, so the tools list is empty. Since it is maintained by the SceneView project rather than Rerun.io, the server is classified as non-official.

Strongest · Maintenance 10/20 Weakest · Reliability 4/20

Reliability
4/20
Security and permissions
5/20
Maintenance
10/20
Documentation
5/20
Setup experience
8/20
Why each score
Reliability 4/20
Only the README and manifest were supplied; no source, tests, or CI evidence for mcp/packages/rerun/server.json was available to verify stdio startup, the MCP init/tool-list handshake, or the real behavior of the claimed 5 tools. The manifest is well-formed and the install command is present, but these are untrusted claims. Under static calibration, missing reproducible evidence must be deducted, so the score is 4.
Security and permissions 5/20
The server description involves streaming AR poses/planes/cloud anchors to a Rerun viewer, Python sidecars, and web viewer embeds, which creates clear external-network and data-boundary risk. No source was provided to confirm user confirmation, least-privilege scoping, or disclosure of data flow. No red-line issue such as malware, credential theft, covert exfiltration, or irreversible operations is evidenced, so the score stays above 4; missing permission/confirmation evidence caps it at 5.
Maintenance 10/20
The repository is not archived, carries an Apache-2.0 license, and shows CI and release activity, so it is broadly maintained. However, there is no server-specific commit history, release cadence, dependency-update evidence, or issue-response evidence for the rerun-3d server, and its version is still 1.0.0. Stars are discovery signals only and do not add points. This matches the “active but governance/versioning gaps” anchor at 10.
Documentation 5/20
The README provides the npx rerun-3d-mcp install command, a claim of 5 tools, and a high-level AR Debug/Rerun viewer workflow, pointing to llms.txt for architecture details. But no reviewable documentation of individual tool parameters, auth, limits, or troubleshooting is included, and the llms.txt content is not in the supplied material. This fits “usable but with hidden assumptions and troubleshooting gaps,” so the score is 5.
Setup experience 8/20
Setup is presented as a single npx rerun-3d-mcp command with stdio transport, so the path is short and clear. However, no client configuration JSON example for this specific server is provided, and there is no executable verification in the files. Static calibration caps setup at 15, and uncertainty lowers the score to 8.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • Developers building SceneView 3D/AR apps (Android, iOS, Web) who already use or want to adopt Rerun.io for debugging.
  • Teams that need replayable AR session evidence for bug reports and post-mortem analysis.
  • Debugging scenarios that combine SceneView AR poses, planes, and cloud anchors with Rerun's timeline visualization.

Not for

  • Projects not using SceneView or Rerun.
  • End users of a released application (this is a debugging tool, not a user-facing feature).
  • Teams that require an officially maintained Rerun.io MCP server and do not accept a community-maintained implementation.

Required permissions

  • No dedicated permission list is published in the source material.
  • Based on the described features, the server may fetch and execute the npm package and its dependencies via npx.
  • It may launch local Python sidecar processes (Rerun SDK).
  • It may write local .rrd recording files and may start or interact with a local web viewer instance.

Risks and side effects

  • The server is community-maintained by SceneView, not by Rerun.io, so standard supply-chain and local-execution risks apply.
  • The npm package and its dependencies may introduce third-party code; pin the version and review the published package before relying on it.
  • The server may process AR session data, so avoid using it with sensitive data in unfamiliar environments.
  • If Python sidecar dependencies are missing, features may fail, and errors may only appear in client logs.

Setup

Before you start

Runtime：Node.js

Run it with npx: npx rerun-3d-mcp. In an MCP-capable client, add a stdio server configuration with command npx and args -y rerun-3d-mcp. The package is downloaded from npm on first run, so network access is required.

claude_desktop_config.json 
{"mcpServers":{"rerun-3d":{"command":"npx","args":["-y","rerun-3d-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": {
    "rerun-3d": {
      "command": "npx",
      "args": [
        "-y",
        "rerun-3d-mcp"
      ]
    }
  }
}

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

Terminal
claude mcp add rerun-3d -- npx -y rerun-3d-mcp

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

Generate and validate client config

Check that it works

Add the server with npx rerun-3d-mcp in your MCP client config and confirm the server starts and its tools appear in the client's tool list; the README does not name individual tools, so successful launch is the check.

Troubleshooting

  1. Verify Node.js/npm is installed and that `npx rerun-3d-mcp` exits cleanly.
  2. Check network access to the npm registry for the first-run package download.
  3. If Python sidecars are involved, confirm Python and the Rerun SDK prerequisites are available.
  4. Inspect the MCP client logs for stderr/stdout output; this is where stdio server errors appear.
  5. Confirm the client configuration uses the exact package name (rerun-3d-mcp) and the stdio transport.

Things to try

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

  • Stream AR debug data from my SceneView session to the Rerun viewer

Use cases

Stream AR session data (poses, planes, cloud anchors) to the Rerun viewer for live inspection.
Record debug recordings (.rrd) and replay them in the hosted Rerun web viewer at sceneview.github.io/rerun/.
Embed Rerun web visualization into spatial-data debugging workflows.
Share a common spatial-data visualization backend across SceneView apps on Android, iOS, and other platforms.

Supported clients

Claude Desktop
Claude Code
Cursor
Windsurf

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

Overview

This stdio MCP server (npm: rerun-3d-mcp, v1.0.0) connects SceneView 3D/AR applications to Rerun.io for spatial debugging. According to the SceneView README, its scope is AR debug logging, Python sidecars, web viewer embeds, and spatial data visualization; the Rerun.io integration is listed with five tools. The server manifest describes it as the Rerun.io integration endpoint within the SceneView repository. It is not maintained by Rerun.io itself.

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 →
Compare with other MCP servers

More servers from this repository

Automotive 3DGaming 3DHealthcare 3DInterior Design 3DSceneview MCP

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