MCP Directory
Back to Blog
ruviewmcp-serverruvnet-ruviewspatial-intelligencewifi-sensing

RuView MCP Server: Real-time Spatial Intelligence from WiFi Signals

November 12, 2026
3 min read

RuView MCP Server: Real-time Spatial Intelligence from WiFi Signals

RuView, a Rust-based MCP Server, leverages standard WiFi signals to generate real-time spatial intelligence, monitor vital signs, and detect presence—all without relying on video. This capability makes it a compelling tool for AI agents that need to understand their physical environment discreetly and non-invasively.

Integrating AI Agents with RuView

RuView provides flexible MCP integration points, supporting both standard I/O and HTTP for agent communication. For general MCP agents, stdio is the default transport. When working with LLMs like ChatGPT, an HTTP interface is available, secured with token protection and restricted to read-only tool access.

To start the RuView MCP server via stdio:

npx @ruvnet/ruview@0.9.1 mcp start

For the HTTP interface, which requires specific grants for device access:

RUVIEW_MCP_GRANTS=device-access npx @ruvnet/ruview@0.9.1 mcp start --http

These commands expose RuView's spatial sensing capabilities to your MCP-compliant agents, allowing them to query and receive real-time data derived from WiFi signals.

RuView's Claude Code & Codex Plugin

A significant aspect of RuView's integration story is its dedicated plugin for Claude Code and Codex. This plugin streamlines the process of onboarding, setting up ESP32 hardware, developing sensing applications, training models, and verifying results. It includes support for advanced sensing, CLI/API/WASM interactions, and mmWave radar.

The plugin provides a live sensing pane at /ruview within Claude Code, shipped directly with the RuView package. To activate this, you can run:

npx @ruvnet/ruview@0.9.1 mod

For a guided setup experience within Claude Code, the plugin can be installed from the RuView repository acting as a marketplace:

# In Claude Code — add this repo as a plugin marketplace, then install:
/plugin marketplace add ruvnet/RuView
/plugin install ruview@ruview

Once installed, the plugin's features cover:

  • Onboarding
  • ESP32 setup
  • Sensing apps
  • Model training
  • Advanced sensing
  • CLI/API/WASM
  • mmWave radar
  • Witness verification

The source for this plugin is located in plugins/ruview/, with its marketplace manifest at .claude-plugin/marketplace.json.

Spatial Sensing Capabilities

RuView's core strength lies in its ability to interpret WiFi signals to infer detailed information about a space and its occupants. This includes:

  • Real-time spatial intelligence: Understanding the layout and dynamics of an environment.
  • Vital sign monitoring: Detecting breathing and heart rates without physical contact.
  • Presence detection: Identifying whether individuals are present in a space.

These capabilities are achieved without any video input, offering a privacy-preserving alternative for environmental awareness. The project also provides pre-trained models, such as wifi-densepose-pretrained, available on Hugging Face, to accelerate development.

References

Back to all posts