
How to run God's Eye View, the open-source spy satellite simulator for your browser
Turn your browser into a live 3D command center tracking real flights, ships, and satellites. Here's how to install God's Eye View for free.
A weekend side project just got the co-founder of Palantir to publicly comment on it. That's not hyperbole — that's what happened when Bilawal Sidhu shared a demo of a tool that fuses live flights, ships, satellites, wildfires, and traffic onto a single photorealistic 3D globe you can run on your own laptop. The project is called God's Eye View, and unlike the billion-dollar intelligence platforms it visually resembles, it's completely free, open source, and installable in about ten minutes.
If you've ever wanted to watch a commercial flight land from the cockpit's point of view, see which container ships are approaching a port right now, or ask an AI assistant to "show me military aircraft over the Pacific" and actually watch the globe respond, this is the tool that does it. Here's what it is, how it works, and how to get it running yourself.
What exactly is God's Eye View?
God's Eye View is an open-source, browser-based application that renders a live, photorealistic 3D globe and layers real-world data on top of it — aircraft, vessels, satellites, active fires, and more. According to the project's own description, it puts all this data in the same place, so you can move between a global picture and an individual aircraft, ship, or street, and it runs locally in your browser with source code you can inspect and extend.
The project started small. A week before going open source, the creator built WorldView — basically a geospatial command center that fuses open-source intelligence feeds onto a 3D globe — as a weekend project, and the internet went kind of crazy over it. The response was intense enough that people started asking for the code, and eventually the creator obliged. To everyone who watched the demos and went off to build their own, and to everyone who kept asking for the code, the response was gratitude — and when polled on whether it should go open source, the answer wasn't subtle.
It's worth noting what this tool is not. The maintainers have drawn a clear ethical line around the project. The project models events, assets, infrastructure, and systems — aircraft, vessels, satellites, fires, cameras, cities — but it does not build features for named-person search, face recognition, or tracking individuals, and pull requests that cross that line won't be merged, because people are not a query type here. That's a deliberate design choice that keeps the tool focused on public infrastructure and events rather than surveillance of individuals.
You can check out the project directly on GitHub, where the full source, documentation, and issue tracker live. There's also a companion newsletter at spatialintelligence.ai where the creator writes about the broader open-source intelligence movement this tool is part of.
What data feeds power the globe?
The magic isn't just the 3D rendering — it's the sheer number of real data sources stitched together into one interface. Community summaries of the project note that God's Eye View renders a live 3D globe with real flights, ships, and satellites pulled from all public sources like OpenSky, CelesTrak, and NASA, with no fake data and nothing paywalled.
Beyond flights and satellites, the tool pulls in environmental and infrastructure data too. One write-up of the project points out that it harnesses an impressive array of live data streams, including aircraft, vessels, satellites, active fires, earthquakes, traffic, public cameras, and physical infrastructure, to create a comprehensive and interactive global overview. Traffic isn't pulled from a live feed the way flights are — instead, according to the repo's own documentation, traffic is simulated along real roads using aggregate location data, while CCTV camera poses and rocket launch trajectories are coarse estimates rather than precise live positions. That's an important distinction: most feeds update in near real time, but a few layers are intentionally approximate, and the project is upfront about which is which.
The interface leans into a few standout features that make exploring this data feel less like a spreadsheet and more like a mission control room:
- Cockpit view — ride inside a tracked flight, with the camera holding the terrain under you all the way down.
- Contacts panel — a 250 km roster of everything near your target, letting you step through live aircraft and drop into any cockpit.
- Click-to-track — the camera locks onto anything you click, draws a fading trail, surfaces full metadata, and a tracked fire or vessel hands you off to the nearest live camera in one click.
Why did this blow up on GitHub Trending?
Part of it is timing — spatial intelligence and open-source data fusion are having a moment. Part of it is the sheer visual punch of watching real ships and planes move across a photorealistic Earth in your browser tab. One analysis of the release called it a project that leverages a photorealistic 3D globe to democratize access to high-fidelity geospatial visualization, representing a shift from high-cost proprietary systems to accessible, open-source web platforms.
That framing matters. Tools like this used to require government contracts or enterprise licensing to access. Now the barrier to entry is a laptop and a free API key. As one write-up put it, the ability to simulate or access a 'spy satellite' perspective was historically reserved for government agencies or well-funded private corporations due to the cost of data acquisition and computational power, and bringing this capability to the browser effectively lowers the barrier to entry.
The project also picked up serious traction on GitHub itself, currently sitting well into the tens of thousands of stars with an active pull request queue from contributors adding new city camera packs and data layers.
How do you install God's Eye View?
Getting it running is more approachable than you'd expect for something this visually dense. The project is built without a heavy framework — it uses no framework at all, just vanilla JavaScript plus CesiumJS and Vite.
Here's the setup process straight from the official README:
- Install the right Node version. You'll need Node.js 24.14.x or 26.x, which is also enforced by the project's package.json.
- Clone the repo and install dependencies. Run:
git clone https://github.com/bilawalsidhu/gods-eye-view.git cd gods-eye-view npm ci npm run doctor npm run dev - Open the local server. Once running, open http://localhost:4173 in your browser.
- Pick a starting mode. You can choose Live Contacts, Space Missions, Environmental, or Explore Manually from the first-run panel.
The best part is that you don't need any API keys just to try it out. No key is required to start — the app boots on keyless Esri World Imagery with keyless terrain, and OpenStreetMap takes over automatically if Esri is unreachable. If you want the full photorealistic 3D globe experience with Google's tiles, you'll need to add a Google Maps API key. You need a Google Maps API key with the Map Tiles API enabled, though most data layers work with no other accounts.
Adding that key doesn't require editing config files by hand. Keys are upgrades, not prerequisites — when you want one, you click the POWER UP chip in the bottom-right corner, and Provider Settings lists every supported provider. Paste your key into that panel and the app writes it to your local environment file automatically.
If you'd rather skip the terminal entirely, there's a no-code path too. People who only want to run the app can install the repository directly through Pinokio, while the terminal path remains available for contributors. You can find the one-click launcher on Pinokio's app page for the project.
Can you control it with your voice?
Yes — and this is where the tool starts feeling less like a map and more like an AI copilot. Voice control lets you speak natural commands and watch the globe respond in real time. Community posts describing the feature note you can say things like "show me military aircraft" or "which ships are heading to Auckland?" or "find the nearest camera," and the AI restructures the map and shows you the relevant objects.
To unlock this, you'll need a separate key. As the same summary notes, launching the app just requires a Google Maps key, while voice control specifically requires an OpenAI key. You add it through the same POWER UP / Provider Settings panel used for the Maps key — no manual config editing required.
Is any of this data authoritative or safe to rely on?
No, and the maintainers are explicit about that. This is an exploratory visualization tool, not a navigation or emergency system. The project's own disclaimer states plainly: God's Eye View is an exploratory visualization of public and third-party data, and that data may be delayed, incomplete, modeled, inferred, or wrong — it should not be used for flight or maritime navigation, emergency response, medical or health decisions, investment decisions, or other safety-critical purposes, and important information should be verified with authoritative sources.
That's a healthy dose of realism for a tool that otherwise looks like something out of a heist movie. Treat it as a way to explore and learn how spatial data fusion works, not as a source of truth for anything mission-critical.
Who maintains the project, and can you contribute?
The project is maintained by its creator alongside a collaborator at a small studio called Halfpixel. Specifically, God's Eye View is maintained by Bilawal Sidhu and Sameh Khamis at Halfpixel, with either maintainer able to review and merge contributions.
Contributions are welcome, but there are ground rules beyond the no-individual-tracking policy. This is a tool for public data — don't add scraping of sources whose terms forbid it, private or paywalled datasets, or anything that misrepresents public-data inference as authoritative intelligence. If you want to add a new data layer, the architecture is built for it: the UI lives in a single file handling panels, HUD, and styles, while layer logic lives in its own per-layer file, and the two are meant to be kept separate.
The whole codebase runs under a permissive license, though the data itself carries different terms. The code is MIT-licensed, but that grant covers the source code only — it does not extend to third-party data or visual assets, since every third-party source keeps its own license and terms.
Should you actually build something with this?
If you're a developer curious about geospatial visualization, real-time data fusion, or just want a genuinely fun weekend project to fork and extend, God's Eye View is one of the more approachable entry points into a space that used to be locked behind enterprise contracts. The README even frames itself as an invitation rather than a finished product: it's described as the canonical live 3D client from the project that kicked off the recent wave of spatial-intelligence tools, and it's a canvas — add a city pack, a data source, a style, a voice tool, because it's the window through which you see the world, and you're invited to bring that window to others.
Clone it, add a free Maps key, and spend twenty minutes watching real flights land from the cockpit view. Then ask yourself what layer you'd add next — because that's exactly the point of the project.