Privacy
Effective November 1, 2026.
The plain version first. Phosphor has no telemetry, no crash reporting, no analytics, no accounts and no ads. I have not built any of that in, first party or third party, and there is no server of mine for the app to talk to. Everything it remembers lives in a folder on your computer. The only things it ever sends go to places you chose yourself: the machine on your network, a Stash server you entered the address of, a page a plugin loads, a Bluetooth device you connected. The one exception is the firmware flasher, which asks openvalence.org for our firmware list when you pick a firmware file. What those endpoints and your operating system do with traffic is outside my control, and this page says where that line is.
The rest is the long version, written so that it is checkable. The source is public; every claim here can be read in the code.
What this covers
This policy covers the Phosphor application on Windows, macOS, Linux and Android, the Nucleus firmware that runs on the machine, and this website, openvalence.org. All of it is published by AtlanticTM under the OpenValence name. "I" below means the one person who maintains it.
What the app stores, and where
Phosphor keeps its state in the per-user application data folder your operating system gives it: %APPDATA% and %LOCALAPPDATA% under com.phosphor.app on Windows, ~/Library/Application Support/com.phosphor.app on macOS, ~/.local/share/com.phosphor.app on Linux (inside ~/.var/app/org.openvalence.Phosphor for the Flatpak), and the app's private storage on Android. The MSIX and installer builds both use these per-user folders. The portable build keeps all its data, including the web view's storage and logs, in a PhosphorData folder beside phosphor.exe, and writes nothing to your user profile or the registry. In the data folder you will find:
- Saved hubs: the address, name and session details of machines you connected to, so the app can reconnect.
- Plugin settings, including, if you set one up, the URL and API key of your Stash server. The key is stored as you typed it, protected only by your operating system's user account, and it is deliberately left out of the app's settings export so it never ends up in a backup file.
- Your themes, layouts, pages, enabled plugins and Neutrino's saved state.
- A log file (
Phosphor.log) with technical events: connections, errors, Neutrino's output. It stays on disk and is never uploaded. If you send it to me for a bug report, that is you sending it. - Media you open in the funscript player is read from where it already is. The app does not copy it anywhere.
Uninstalling removes the application. The data folder may stay behind depending on the platform; deleting it deletes everything the app ever knew about you.
What the app sends, and to whom
Every network connection Phosphor makes goes to an endpoint you chose, apart from the firmware list request below. There is no endpoint of mine.
- Your machine
- To find a hub, the app sends a short discovery broadcast on your local network and listens for replies; it only broadcasts when you ask it to scan, or when it reconnects to a hub you saved. Once connected, the app and the hub exchange machine state over a WebSocket on your network. An emergency stop is also sent as a broadcast so it reaches the hub even if the session has dropped.
- Stash
- If you enter a Stash server in the funscript player, the app sends requests to that address with the API key you gave it, and streams video from it. That is your server, on whatever network you put it on; its own logs are its business.
- Plugins and the pages they load
- The app allows plugins to fetch
httpandhttpsURLs. The plugins that ship with the app only contact addresses you configured. A plugin you install from elsewhere can contact whatever its author wrote it to; read its manifest before enabling it. - Other programs on your computer
- Phosphor runs a Buttplug server so that other software can drive the machine and your toys through it. It listens on the loopback address only (
127.0.0.1, port 12345 by default), so only programs on the same computer can reach it. It never listens on your network unless you change that setting. - Firmware updates
- When you pick a firmware file, the app asks openvalence.org for our firmware list (flash/fw/index.json) to label it Official or Unrecognized. Nothing about the file or your machine is sent; the comparison runs on your computer. The file goes only to your machine.
- Neutrino, the built-in hub
- It runs inside the app, as WebAssembly, and makes no network connections of its own.
The app does not check for updates by itself. Updates arrive through whatever installed it: Flathub, or a download from GitHub.
Bluetooth
Phosphor uses Bluetooth Low Energy for two things: finding and talking to a hub when Wi-Fi is not available, and, through ButtplugIO, finding and controlling toys. A scan runs when you press Scan, and the app connects only to the device you pick. It reads the advertising data the devices around you broadcast during a scan, keeps it in memory for the list you see, and discards it when the scan ends; it does not keep a record of devices you did not connect to.
On Android, the operating system requires the location permission before any app may scan for Bluetooth devices, because scan results can in principle be used to infer location. Phosphor asks for that permission for that reason only. It does not read your position and has no code that would.
The machine
Nucleus, the firmware on the hub, stores the machine's configuration and the names of the clients that are allowed to control it, on the board itself. It has no internet connection of its own and no server it reports to. It answers discovery on your local network and serves the clients that connect to it. Resetting the board's configuration clears it.
What is outside my control
Parties that may learn something about you while you use this software. I run none of them and get no data from them:
- Your operating system
- Windows, macOS, Android and Linux distributions collect their own diagnostics about the applications you run, under their own settings and policies. On Windows the app renders through WebView2, which is Microsoft's component and follows Microsoft's telemetry settings for your account.
- Where you got the app
- Flathub and GitHub each count installs and downloads and may show me aggregate numbers. They do not give me who you are.
- Addresses you configured
- Your Stash server, any URL a plugin loads, and the hub itself all see the requests the app makes to them.
- Bluetooth devices
- A connected device learns that something connected to it. What a toy's manufacturer does with that, if it has its own app or cloud, is theirs to disclose.
This website
openvalence.org is static files. It sets no cookies and has no ads or third-party trackers. If you pick a theme with the sun/moon toggle, that choice is saved in your browser's local storage. Its fonts are served from its own folder, not a third party. The downloads page asks GitHub's API (api.github.com) for the newest Phosphor release each time it loads, so GitHub sees that request, as it sees the download itself. The documentation at /Valence/ and the playground at /Kinetic/ are the same kind of static site.
The motor setup page, /guides/servo-setup/, talks only to the serial port you choose in the browser's Web Serial prompt. The motor's registers and the frame log stay in your browser.
The web flasher, /flash/, fetches firmware from this site only: the list of releases when the page loads, and a release's files when you pick it. Image files you pick from your computer are read in the browser. Either way the images go only to the serial port you choose. The files, the board's details and the session log stay in your browser and are not sent anywhere. A Wi-Fi network you type into the flasher goes only into the board's settings storage over that port: the page does not store it, and the log records only the length of the network name.
Visitor statistics
Visits are counted with Umami, an open-source analytics tool I host myself, at home in the United States; its only public address is stats.openvalence.org. No analytics company is involved. I use it to see which pages get read and in what order. The data stays on that server: not sold, not shared, not used for ads.
- Recorded per page view: page address, referring page, browser, operating system, device type, screen size, language, country and region.
- IP address: used to derive country and region, then discarded. Never written to the database.
- Visitor ID: a one-way hash of IP, browser and site, salted with a secret that rotates monthly. It can't link you across months, across sites, or back to you.
- Your browser: the statistics script sets no cookies and writes nothing to local storage. No session recordings, no heatmaps. It reads one opt-out key.
- Opt out: configure your browser to send Do Not Track, block the script with a content blocker, or set
umami.disabledto1in this site's local storage. Any of the three stops records from being created.
EU visitors
Built for the GDPR and ePrivacy rules. The site writes one thing to your device: the theme you pick with the theme button, kept as theme (light or dark) in local storage. That is storage you asked for, which ePrivacy Article 5(3) exempts from consent. Nothing else is written, so there is no cookie banner. Deriving country and region from your IP rests on legitimate interest (GDPR Article 6(1)(f)), with the minimum data needed. Records hold no identifier that links back to you, so there's no per-person record to look up, export or delete. Anything else: write to [email protected].
Who serves it
GitHub Pages hosts the files and Cloudflare runs the domain. Both may keep standard server logs (IP address, pages requested, browser) under their own policies: GitHub's privacy statement and Cloudflare's privacy policy. I don't have access to those logs.
Google Search Console
I read Google Search Console for this site. That is Google's own data about how the site shows up in Google Search: which queries it appeared for, how often, how many clicks, which pages. Google collects and aggregates it from its search results; the site runs no Google code to produce it. See Google's privacy policy.
Changes and contact
If any of the above changes, this page changes with it and the date at the top moves. The page is a file in a public repository, so its history is the changelog: every revision.
Questions, or something here that does not match what the code does: [email protected]. If the code and this page disagree, the code is right and this page gets fixed.