Privacy

Effective: October 27, 2026 · Last updated: October 27, 2026

What changed: added anonymous website page counts (section 4): no cookies, no stored identifiers, kept 90 days; also the aggregate daily usage counts of the API/MCP service, the hashed rate-limit counters, how long they are kept, your rights and our legal basis, and our cookie wording.

In short: there are no accounts. Your music isn't stored on our servers. It stays in your browser, in files you save, or in links you share. The one exception is the temporary audio files made when an assistant or program uses the API/MCP render service (section 4). This page lists everything that does touch servers.

1. Who we are

Moose Tracker is run by Kurtis Alexander. Contact: moose@moosetracker.app.

2. What we do not collect

No accounts, sign-in, names, email addresses, or profiles.
No songs, instruments, samples, recordings or settings. The website never sends them to us. (The API/MCP render service in section 4 is the one exception.)
No advertising, no tracking cookies, no third-party analytics scripts and no sale or sharing of personal information. The website sets no cookies of its own. Cloudflare, which delivers the site, may set a strictly necessary security cookie (for example, to tell bots from people) if its bot protection is switched on; if that happens we will list it on the Local storage page.

3. Where your work lives

In your browser: songs, your sound library, autosave and settings are stored in your browser's local storage and IndexedDB on your device. Clear them any time (how).
In files: .mtk files, WAV, MP3 and Opus exports that you save are on your device.
In share links: a share link carries the song in the part of the address after #. Browsers do not send that part to the server, but it travels wherever you paste the link, and anyone with the link can open the song. Share links are not private: do not put anything secret in a song.

4. What does touch servers

Website hosting. The site is delivered by Cloudflare. Like any web host, Cloudflare processes technical request data (such as IP address, user agent, requested file and time) to serve pages and to protect against abuse, and may keep security and operational logs according to its own policies. We do not use this to identify you.
Rate limiting (API/MCP service only). To prevent abuse, our server computes a salted one-way hash of a caller identifier (the anonymous ID that an assistant platform supplies, or, if there is none, the caller's IP address) and keeps counters against that hash: calls per minute, renders per hour and per day. These counters hold no music, expire automatically within about 2 days, and are used only to enforce limits. We do not store the raw IP address. Cloudflare itself sees IP addresses as described above. We do not use this data to identify, profile, advertise to, or contact anyone.
Aggregate usage counts (API/MCP service). To see whether the service is used and working, the server keeps daily totals as plain numbers: how many calls each tool received (describe, sfx, song, render, convert), how many were check-only, how many audio files were made in each format (WAV or Ogg), and how many calls ended in an error or were slowed by a rate limit. It also keeps the number of different callers each day, how many of them were new, and how many had also called within the previous 30 days. To count callers without knowing who they are, the caller identifier is hashed again with a secret salt that includes the date; that list is deleted when the day ends. A second salted hash, kept only with the date it was last seen, is used for the 30-day "returning" count and is deleted after 30 days. The totals contain no song text, prompts, IP addresses, or user identifiers, and are not used to identify, profile, or contact anyone or for advertising.
Website page counts (anonymous). The website adds up visits as plain daily totals: which page was opened (for example the editor or this page); the name of the site that sent you, if any (the site name only, never the full address); a broad device type (phone, tablet, computer or car), worked out from your browser's description and then discarded; whether a visit lasted about 30 seconds or you pressed play, edited or used the keys; whether it went on for about 5 minutes or used several features; and how many shared-song links were opened (never the song). There are no cookies and no stored identifiers. To count how many different visitors there are each day, your network address and browser description are combined with a secret salt and the date and one-way hashed; that list is deleted when the day ends, and we do not recognise returning visitors. These totals are kept for 90 days and are not used to identify, profile, advertise to or contact anyone. Nothing is counted if your browser sends Do Not Track or Global Privacy Control.
API / MCP requests. When you or an AI assistant you use calls our API or MCP service, the song text in that request is processed in memory to answer it. We do not keep the text. A request for audio is passed to our rendering service, a cloud function that produces the audio file.
Rendered files (only when a render is requested). An audio file made by the render service is placed in temporary storage and is deleted automatically after about 2 hours (a storage rule removes anything left over after 1 day). Anyone with the download link can fetch the file until then, so treat links as private, and download anything you need to keep. If the render service is not switched on, no such files exist.

5. Service providers (subprocessors)

We use Cloudflare for hosting, security, rate limiting and temporary file storage, and Amazon Web Services (AWS Lambda) for the render function that creates audio files. These providers process data on our behalf under their own terms and privacy policies. We do not share data with anyone else, and we do not sell it.

6. Your rights (including EEA, UK and Switzerland)

For the small amount of data described in section 4 (the hashed counters), our legal basis under the GDPR and UK GDPR is legitimate interests, Article 6(1)(f): keeping a free service available and preventing abuse. Our providers are Cloudflare (hosting, Workers, R2 storage) and Amazon Web Services (the render function). They act for us under their data processing terms, and where data moves to the United States they rely on the EU-US Data Privacy Framework and standard contractual clauses. Because the counters are one-way hashes with a secret salt, we cannot tell whose they are unless you give us an identifier and we can recompute the hash; if you do, we will tell you what we hold or delete it. You can object, and you have the right to complain to your data protection authority. California and other US state residents: we do not sell or share personal information, we are below the thresholds of those laws, and we treat access and deletion requests the same way. To make a request, write to moose@moosetracker.app.

7. Children

Moose Tracker is a general-audience music tool. It has no accounts, chat or public posting and does not knowingly collect personal information from children under 13 (or under the age set by local law). The only identifier used is the network address handled by Cloudflare and, for the optional API/MCP service and the anonymous daily visitor count, a one-way hashed form of it, used only to secure and operate the service and to limit rates (support for internal operations under 16 C.F.R. § 312.2). It is not used to contact anyone, build a profile, or show advertising. Parents and guardians: because the website keeps your work on the device, there is no profile of your child on our systems. If you believe we hold personal information about a child, write to moose@moosetracker.app and we will delete it.

8. Your choices

Because the website holds no account and no copy of your music (rendered audio files aside, which are deleted within about 2 hours), there is nothing to download or delete on our side. You can delete local data in your browser at any time. If you believe we hold other personal data about you (for example in something you emailed to us), contact moose@moosetracker.app; we will respond and, where appropriate, delete it. Depending on where you live, you may have additional rights under local law (see section 6); contact us to exercise them.

9. Security

Data transfers use HTTPS. No online service is perfectly secure and we cannot guarantee absolute security. To report a security problem, write to moose@moosetracker.app (see also security.txt).

10. Changes

If we change what we collect, we will update this page and its date, and note the change at the top. Changes apply from the date they are posted.

11. Contact

moose@moosetracker.app

Contact: moose@moosetracker.app.