Skip to main content
Glama

livetrafficcam-mcp

MCP (Model Context Protocol) server for LiveTrafficCam, a directory of live US traffic cameras built on official state DOT feeds. Every camera is checked on a rolling schedule with real HTTP requests, so "live" means a verified current image, not a listing.

The server wraps the site's public JSON API. It sends one request per tool call, caches responses for 60 seconds, and returns compact JSON with a source_url you can cite.

Install

Requires Node 20 or newer.

Claude Desktop (claude_desktop_config.json):

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

Claude Code:

claude mcp add livetrafficcam -- npx -y livetrafficcam-mcp

Cursor (.cursor/mcp.json):

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

Related MCP server: Vancam MCP Server

Tools

  • list_states — states covered, camera counts, and a liveness summary per state. Example: "Which states have traffic camera coverage?"

  • cameras_by_state — cameras for a two-letter state code, with an optional route filter. Example: "Show live cameras on I-90 in Washington."

  • cameras_for_entity — cameras for a named pass, corridor, bridge or tunnel slug. Example: "What do the Donner Pass cameras show?" uses donner-pass.

  • camera_status — one camera's verified-live status, last live time, and official DOT source. Example: "Is the I-90 floating bridge camera up?"

  • state_uptime — measured uptime for a state's camera fleet: live, stale and dead counts plus the 14-day check success rate. Example: "How reliable are Washington's traffic cameras?"

The camera API returns at most 200 cameras per query; tool output says so when a state has more. Fleet counts come from the camera uptime report.

Data source and attribution

All data comes from livetrafficcam.com's public API, which aggregates official state DOT camera feeds (WSDOT, Caltrans, UDOT, and others). Every camera result carries the attribution line for its agency; keep it when you display the data. Uptime numbers should cite the uptime report.

More from the same data:

Also available

A Home Assistant integration puts a pass, bridge, tunnel, airport or highway on a dashboard: one camera entity per camera, a verified-live sensor each, and reported disruptions. No API keys.

License

MIT

Available Tools

5 tools
cameras_by_stateA

List traffic cameras for a US state (two-letter code), optionally filtered by route (e.g. "I-80"). The API returns at most 200 cameras per query; use state_uptime for full fleet counts.

Inputs: state (required), route, live_only.

cameras_for_entityA

List traffic cameras for a named pass, corridor, bridge or tunnel by its slug (e.g. donner-pass, snoqualmie-pass). Includes the matching livetrafficcam.com page URL.

Inputs: entity_slug (required).

camera_statusA

Live status of one camera: verified-live state, last live time, official DOT source URL and attribution. Identify the camera by id, slug, or a name fragment, within a state.

Inputs: state (required), name, slug, camera_id.

list_statesA

List the US states covered by livetrafficcam.com with camera counts and a liveness summary per state. Counts come from the measured camera uptime report.

Inputs: no named parameters.

state_uptimeA

Measured camera uptime for one state: how many official DOT cameras are live, stale or dead right now, and the 14-day check success rate. Cite the report URL when quoting these numbers.

Inputs: state (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.8/5.0

    Scored across 5 tools

    Disambiguation4/5

    Each tool targets a distinct lookup type: state overview, cameras by state/route, cameras by named entity, single-camera status, and uptime statistics. There is minor overlap between list_states and state_uptime since both mention liveness summaries, but their levels of detail and purposes are clearly different.

    Naming Consistency3/5

    Naming is readable but inconsistent: list_states uses a verb, cameras_by_state and cameras_for_entity follow a noun-by-entity pattern, while camera_status and state_uptime are noun-noun phrases. The conventions are mixed but still predictable enough to navigate.

    Tool Count5/5

    Five tools is well-scoped for a traffic camera lookup server. Each tool covers a necessary access pattern without redundancy, and the count feels neither thin nor bloated.

    Completeness4/5

    The set covers the core domain well: state coverage, camera listing by state/route and named entities, single-camera status, and uptime reporting. Minor gaps exist, such as no direct camera-image or stream endpoint and no free-text national camera search, but agents can work around these with the provided URLs.

    Maintenance

    ActivityStale
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      C
      maintenance
      Provides real-time access to Swedish road weather, traffic cameras, road conditions, and traffic flow data from Trafikverket (Swedish Transport Administration). Enables users to query weather stations, cameras, traffic situations, and flow data through natural language.
      10
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Provides access to Iowa DOT driver license station data, including hours, CDL testing info, and live queue camera images from waiting rooms.
      64 npm
      MIT
    • A
      license
      A
      quality
      B
      maintenance
      Provides live California road conditions, route planning, and an AI assistant over MCP, enabling natural-language queries about traffic, closures, chain controls, and more.
      18
      6
      MIT