1. Who we are, and two different roles
Multiplayer Agent is owned and operated by Midnight Monsters Corporation, which is the data controller responsible for the personal data described here. “We”, “us”, and “our” mean Midnight Monsters Corporation throughout. Reach us about anything on this page at info@midnightmonsters.co.
We handle two kinds of data, and the rules differ for each.
- Your data, as our customer. Your studio account, your games, your billing. Here we are the controller: we decide what to collect and this policy governs it.
- Your players’ data. The identities, saves, scores, and gameplay analytics of people who play games built on our platform. Here we are a processor acting for the studio that made the game: that studio decides what its game collects, and its own privacy notice governs. We process that data to run the service and on the studio’s instructions.
If you are a player and want your data removed from a game, the fastest route is the studio that publishes it. You can also write to us at info@midnightmonsters.co and we will route it.
2. What we collect from studios
- Account. Your email address, a hashed password (we never store the password itself), your plan, the access code you signed up with, the members you invite to your studio, and account timestamps.
- Your games. Game source code and assets, the prompts and instructions you give the builder or an AI agent, build and revision history, room configuration and server-side game logic, and the runtime logs those produce. This is your Confidential Information: see Section 7 of the Terms of Service.
- Billing. Your subscription and credit balance, an identifier from our payment processor, and invoice history. Card numbers go directly to Stripe; we never receive or store them.
- Support. Bug reports, error traces, and anything you send us when you ask for help.
- Operational logs. Requests to our API and realtime servers, with the usual server metadata (timestamp, endpoint, key used, response status, coarse error detail), kept for debugging, abuse prevention, and billing accuracy.
3. What the platform collects about players
These are the fields the platform itself records when a game uses our SDK. A studio can turn features off, and a studio can also add its own collection that we know nothing about.
- A player identifier. A random device-scoped id stored in the browser’s local storage. It is not your name, email, or account: it exists so a player keeps their progress between visits.
- Optional claimed accounts. If a game lets players claim an account, we store a username and a hashed password. If a game uses a portal login (for example a game portal that hosts it), we store the identifier that portal gives us, not the portal password.
- Gameplay data. Saved progress, leaderboard scores, entitlements, and any content the player creates and publishes through the game’s content features.
- Analytics events. Event names and a small set of properties, the room and session involved, the page host and referrer, the browser’s reported timezone, and the browser user-agent string.
- A hashed IP address. We do not store raw IP addresses in analytics. We store a salted one-way hash, used to tell distinct devices apart and to catch abuse. It is not reversible into an address.
- Realtime session records. Server-side connection records, when a session joined and left which room used for concurrency limits, billing accuracy, and the analytics we show studios.
4. What we deliberately do not do
- No third-party analytics or trackers. Our analytics are entirely first-party, written by us and stored in our own database. We do not embed Google Analytics, Meta pixels, or any comparable tag in the SDK or the dashboard.
- No advertising network of ours. We run no ad network and place no ad tags. A studio may add its own ad provider to its own game; if it does, that provider’s privacy practices apply to that game, not ours.
- No selling or renting data. Not yours, not your players’. We do not share it for cross-context behavioural advertising, which is what “sale” or “sharing” means under California law.
- No training on your games. We do not use your game code, your prompts, or your build history to train models, and we do not hand them to a model provider for training.
- No special-category data by design. The platform is not built to hold health, biometric, precise location, or government-identifier data, and studios agree not to send it.
5. Why we process it
- To provide the service: run rooms, store saves, build and host games, show analytics. (Under GDPR: performance of a contract.)
- To keep it working and safe: debugging, rate limiting, abuse and fraud prevention, capacity planning. (Legitimate interests.)
- To bill accurately: metering usage and credits. (Contract; legal obligation for tax records.)
- To support you: answering your messages and investigating what you report. (Contract; legitimate interests.)
- To improve the platform: aggregate, non-identifying usage patterns across the service. (Legitimate interests.) This never involves showing your game to anyone or using it to build a competing one.
6. Who else sees it
We use a small number of infrastructure providers, each under a contract that requires them to protect the data and use it only to provide their service to us:
- Railway: application hosting and the managed Postgres database where account, game, and analytics data live.
- Anthropic: the AI models behind the builder and the MCP tools. Prompts and the game context needed to answer them are sent to the model provider to generate a response. See Section 6 of the Terms for what that means for ownership of the result.
- Stripe: payment processing and card handling. Stripe is the controller of the card data it collects.
- Game portals: only where a studio enables a portal login for its game, and only the identifier needed to sign that player in.
Beyond that, we disclose data only when the law compels it, when it is necessary to investigate a security incident or abuse, or when you tell us to. If we are ever acquired or merge, data moves with the business and stays subject to this policy and to our confidentiality obligations; we will tell you before that happens.
7. Local storage and cookies
The dashboard keeps your login session token in your browser’s local storage rather than in a cookie. Games using our SDK keep a player identifier and, if the player has one, a player session token in local storage and fall back to session storage in embedded contexts where local storage is blocked. None of this is used for advertising or cross-site tracking, and we set no third-party cookies.
8. How long we keep it
- Account, game, and player data, for as long as your account is open, and then deleted or anonymized within a reasonable period after closure.
- Analytics events: retained to power the dashboard’s historical charts; older raw events are pruned or aggregated as the dataset grows.
- Operational logs: a short window, sufficient for debugging and abuse investigation.
- Billing records, as long as tax and accounting law requires, typically several years.
- Backups: deletions propagate through backups on their normal rotation rather than instantly.
9. Your rights
Depending on where you live, you may have rights to access, correct, export, delete, or restrict processing of your personal data, to object to processing based on legitimate interests, and to complain to your data protection authority. California residents have equivalent rights, including the right not to be discriminated against for exercising them, and, as noted above, we do not sell or share personal information.
Most of this is self-serve: the dashboard lets you export your analytics events and your account data, and delete games and players. For anything else, write to info@midnightmonsters.co and we will respond within the time your law requires, and in any event without undue delay.
10. Children
Studio accounts are not for children: you must be at least 13, and 16 where local law requires it. Games built on the platform may be played by children, and the studio publishing a game is responsible for complying with children’s-privacy law such as COPPA. We do not knowingly collect personal information from a child beyond the anonymous identifiers described above; if you believe we hold a child’s personal data, tell us at info@midnightmonsters.co and we will delete it.
11. Security
Traffic to our API and realtime servers is encrypted in transit. Passwords are stored only as salted hashes. Player-facing IP addresses are stored only as salted hashes in analytics. Server-side game logic submitted by studios runs in an isolated sandbox so one customer’s code cannot reach another’s data. Internal access to customer data is limited to personnel who need it to run or support the service. No system is perfectly secure; if a breach affects your data we will notify you as the law requires and as quickly as we can.
12. Where data is processed
Our infrastructure and our providers are located primarily in the United States, so data you send us is processed there. If you are in the European Economic Area, the United Kingdom, or Switzerland, we rely on appropriate safeguards: standard contractual clauses with our providers, for those transfers.
13. Changes
We will update this policy as the platform changes. Material changes get notice in the service or by email before they take effect, and the “last updated” date above changes with every revision.
14. Contact
Privacy questions, data requests, or a suspected incident: write to Midnight Monsters Corporation at info@midnightmonsters.co.