ProjectSkyGrid

Browser-First Airspace Monitoring

For operations teams, analysts, and enterprise programs

Detect and review aircraft anomalies without leaving the browser.

SkyGrid helps analysts and operations teams monitor live traffic, triage suspicious patterns, and forward validated events into downstream workflows with structured data.

ConnectingFetching live engine summary

HERO Monitor Widget

Live operational pulse

Engine

Connecting

Fetching live engine summary

Tracked Aircraft

Syncing...

Current monitored aircraft footprint

Active Anomalies

0

Current anomaly events in active feed

Alerts (24h)

0

Delivered anomaly alerts over last 24 hours

Live Monitoring Surface

Watch-area map and event layers

Initializing globe renderer...

Loading Cesium 3D engine

Detailed historical aircraft leaderboard views are available after sign-in to avoid exposing full event history publicly.

Rollout Essentials

Everything needed to validate and deploy confidently

Detection criteria (current)

  • Loiter: aircraft remains in the same H3 cell beyond the configured dwell threshold.
  • Ghost: missing callsign with non-zero position and high-altitude profile (clean-quality filtered).
  • Squawk: emergency transponder codes 7500, 7600, and 7700 with transient commercial suppression.
  • Quality mode: clean applies false-positive suppression; raw preserves unfiltered event history.

Event classifications are advisory and intended for analyst review. They are not final determinations.

Data provenance and freshness

SkyGrid combines ADS-B telemetry with contextual enrichment (weather, geocoding, NOTAM). Stream and summary health indicators on this page expose current availability and degradation states in real time.

Sample Event Payload

{
  "anomalyType": "rapid-descent",
  "aircraftIcao24": "redacted-icao24",
  "callsign": "REDACTED01",
  "latitude": 40.7128,
  "longitude": -74.0060,
  "quality": "clean",
  "observedAt": "2026-04-21T14:06:12Z"
}

Next Step

Move from live visibility to faster analyst response.

Start with one watch area, tune a focused rule set, and validate event delivery before scaling coverage.