Privacy

Privacy policy.

Effective date: 9 October 2026

1 Who operates hinky.dev

More On Off Switches LLC, doing business as hinky.dev (“we,” “us,” or “our”), operates hinky.dev and its browser-based integrated development environment (the “Service”). We operate from Texas, United States.

This Policy describes how we handle personal information about visitors, account holders, and people who contact us. Source code and other project material may also contain personal information.

Contact: privacy@hinky.dev. Mailing address: 5900 Balcones Drive, Suite #7984, Austin, TX 78731.

2 The Service and anonymous access

The Service provides browser-based editing, building, running and tracing of code. An unregistered session can do all of that, within limits described in our Terms: its network reaches only the package sources a build needs, nothing about it is reachable from outside, and it stops after five minutes. “Anonymous” means no account is required; it does not mean we receive no identifying information. Server requests expose technical information such as IP addresses, and running your code on our servers requires processing it.

We do not currently accept payments or collect payment-card information, and everything the Service offers is free of charge. If we introduce paid plans we will update this Policy before processing any payment.

3 Information we process

Depending on how you use the Service, we process:

Account and authentication information. An account is held by our authentication provider. You can create one with an email address and a password, or sign in through Google. We keep an account identifier, authentication records such as sign-in times, and your preferences. There is no registration form beyond that.

If you sign in through Google, we receive from Google the identifier it holds for you, your email address, and your first and last name where its response includes them. We never receive the password for that account. We ask Google for nothing else: not your mail, your files, your contacts, your calendar, or anything beyond confirming who you are. Signing in this way also tells Google you are using hinky.dev, which is unavoidable because your browser visits Google to sign you in. If you would rather it did not know, use an email address and a password instead.
Project content, including submitted code, files, configuration, dependencies, build artifacts, and compiler output.
The editing history of your project content. The Service records your edits as you make them, so that your work survives a closed browser and follows you between browsers. That record is more granular than a sequence of saved versions: it is the stream of changes itself, and it is retained alongside the current state of each file. It is the largest body of information we hold about how you work, and it is subject to the same retention rules, deletion requests, and access restrictions as the rest of your project content.
Technical and operational information, including IP address, browser information, session identifiers, request timestamps, errors, job status, and resource usage.
Security information, including abuse indicators, enforcement records, complaints, and information relevant to investigations.
Communications you send us, including support requests and privacy requests.
Early-access requests. If you ask for early access from our landing page, we keep the address you submit together with the IP address, browser user agent, and time of the request. This happens before any account exists, and the record is kept separately from one.

We do not collect date of birth, age, or any age-verification record. Section 12 explains how the age requirement works without one.

Information comes from you, your browser and use of the Service, and any authentication or other service you choose to connect. Do not submit unnecessary sensitive personal information or credentials in project files or support messages.

4 Purposes and legal grounds

We process information to provide requested editing and build services, authenticate users, preserve saved work, troubleshoot faults, enforce resource limits, investigate complaints, prevent abuse, protect our infrastructure, respond to requests, and comply with law.

Where GDPR or UK GDPR applies, we rely on contract performance for processing necessary to provide a service under a valid contract; legitimate interests for proportionate security, abuse prevention, reliability, and support, subject to an assessment of your rights and interests; legal obligations for required records and disclosures; and consent where required for optional processing. We take particular account of children’s interests. We do not use a different legal basis merely to avoid a required parental-consent process.

Providing information needed to deliver a feature may be necessary to use that feature. Optional information is not required for unrelated features.

We do not sell personal information, share it for cross-context behavioral advertising, use it for targeted advertising, or use private code to train general-purpose AI models. We use no third-party analytics or advertising services at all, so there is no mechanism by which such sharing could occur inadvertently through a vendor's default settings. We do not make solely automated decisions with legal or similarly significant effects. Automated security controls may block or limit jobs; you can contact us to request review.

5 Private sessions and operator inspection

Sessions and projects are private from other users. Submitting code does not make it public. However, code and related material processed or stored on infrastructure operated for hinky.dev can be accessed by the operator and authorized providers.

We may inspect and analyze code, files, artifacts, and logs through automated systems and manual review where reasonably necessary to operate and troubleshoot the Service, investigate complaints, detect or prevent abuse, enforce our Terms, protect users and infrastructure, or comply with law. We do not request separate permission for each review.

Access is limited to people and providers with an operational need and appropriate access and confidentiality obligations. Review is limited to these purposes; it does not authorize unrelated publication or commercial exploitation of private projects. A private session is not a promise that the operator cannot access its content.

6 Browser storage and third-party AI

Saved source code can persist across browser exits. Closing the browser does not itself delete saved work. Browser copies and copies on our infrastructure have separate retention and deletion processes.

AI API keys are stored only in your browser. We never save a key to our databases, and a key is never written to persistent storage on our servers. A key is, however, transmitted through our infrastructure when you use an AI feature, because the feature runs in a build environment on a server rather than in your browser: the key travels with your request over the same connection the editor uses, and is supplied to that environment as a configuration value for the life of the session. While the session runs, the key is therefore present on the machine running it. It is removed when the session ends.

This means the honest summary is that we receive your key in order to pass it to the environment that needs it, and we do not retain it. If you prefer that a key never leave your browser at all, do not use the AI features.

The same path carries the rest of the interaction. Your prompt is sent with the request; the AI feature reads the project files in that environment; and the provider's responses return over the same connection to your browser. Prompts, project content, and responses therefore pass through our infrastructure in transit. They are not stored for AI purposes, and the operational logging described in section 8 applies to them as it does to other build activity.

hinky.dev does not provide native AI tools or an included AI subscription. You are responsible for the third-party AI services you choose to use, their eligibility rules, and their charges. An AI provider receives the information you send to it under its own terms and privacy policy.

We use the following browser storage, all of it necessary to operate the Service rather than optional:

Local storage for your preferences and, for an unregistered session, the credential that identifies that session.
IndexedDB for your project files and their edit state. This is what allows saved work to survive closing the browser; it is a copy held on your device, separate from the copy on our servers.
Cookies set by our authentication provider on our sign-in domain, for the sign-in process itself. If you sign in through Google, Google sets its own cookies on its own sign-in pages, under its own policy; we neither set nor read them.

We use no third-party analytics. There is no Google Analytics, no advertising tracker, and no behavioural or session-recording tool of any kind. We removed Google Analytics on 6 October 2026 and have not replaced it.

No third party is contacted when you load the Service. Fonts, scripts, and styles are served from our own domain, so loading the editor does not tell any other company that you visited us or reveal your IP address to them. A request leaves our domain only when you direct it to: connecting a code repository reaches that provider, and using an AI feature with your own key reaches that provider, as described above.

Because every item of browser storage listed above is necessary to provide a service you asked for, and because we set nothing optional, we do not ask you for cookie consent and have no consent banner. That is a consequence of collecting less, not an exemption we have claimed for tracking: if we later add anything that is not strictly necessary, we will ask first and this section will say so.

What we do see is what any web server sees: our content delivery network records requests, including IP address, time, the resource requested, and your browser's user agent. Section 8 gives the retention period for those logs. That information stays with us and our hosting provider; it is not shared with an analytics company.

You can clear browser data through your browser settings, although this may remove saved work or prevent features from operating.

We do not change our necessary processing in response to a browser Do Not Track signal. Because we do not sell or share information for targeted advertising, no such activity occurs regardless of whether Global Privacy Control is enabled. We honor legally required opt-out signals if our practices change.

7 Recipients and providers

We disclose information to providers supporting hosting, storage, network delivery, authentication, security, monitoring, and support, only as needed for their functions and subject to appropriate contractual restrictions. Our current providers and their roles are:

Amazon Web Services — all hosting, storage, content delivery, authentication and outbound email. Your projects, their editing history and our logs are held here. This is the only provider that holds your project content.
Google — Workspace is the mail service behind our @hinky.dev addresses, so it receives email you send us; and, if you choose it, Google is the identity provider that signs you in.
GitHub — only if you connect a repository, and only what that requires.
Anthropic — only if you use an AI feature with your own key, as section 6 describes.

That is the complete list. We use no analytics provider, no advertising provider, and no third-party service that your browser contacts when it loads the Service.

We may disclose information to a service you direct us to connect to; to advisers, authorities, or other recipients where reasonably necessary for legal obligations, valid legal process, security, or legal claims; or to parties involved in a business transaction under appropriate safeguards. We provide legally required notice of material changes associated with such a transaction.

An independently selected AI provider is governed by your arrangement with it; bring-your-own-key usage does not automatically make that provider our processor. Google, when you sign in through it, is likewise responsible for its own handling of that sign-in under its own policy; what it tells us is listed in section 3.

8 Retention and deletion

Information	Retention rule
Browser-stored code and AI keys	Until you remove them, clear site data, or the browser removes them; browser persistence is not guaranteed.
Temporary server build files and exports	Deleted within 24 hours after a job finishes or is terminated, unless expressly saved as a project. Execution traces have their own row below, because they are kept longer.
Server-saved anonymous projects	Deleted 7 days after the session begins; the expiry is displayed in the Service while the session is open.
Registered account projects and account profile	Kept while the account remains active; deleted from active systems within 30 days of a verified deletion request or account closure.
Inactive registered accounts and projects	After 12 months without use, we send 30 days’ advance notice to the account email before deletion. Signing in during that notice period cancels inactivity deletion.
Execution traces you record	Up to 30 days from creation.
Logs of every kind	No log we keep retains information for more than 30 days. This is a ceiling rather than a list, and it covers our content delivery network's request logs (which include your IP address), our application and function logs, and our API's own logs. It covers any log not named here as well, which is deliberate: a promise that only covers the logs someone remembered to enumerate is not a promise.
Security and abuse logs	Up to 30 days from creation, except relevant incident evidence below.
Early-access requests	Kept until you ask us to remove the request, or until the account created from it is closed; there is no automatic expiry. Write to privacy@hinky.dev to have one removed without creating an account.
Support correspondence and privacy-request records	Up to 12 months after resolution, excluding attached code that is no longer needed.
Backup copies	We do not keep backups. See the note below the table.

No backups, and no recovery. We do not maintain backup or archival copies of your projects, and we have no mechanism to restore work that has been deleted, overwritten, or lost to a fault on our side. Deletion is permanent, whether you asked for it or it happened by accident. This is a deliberate statement of our current capability rather than a limitation of liability: if your work matters to you, keep your own copy.

How to keep your own copy, and the current limits on it. A registered account can export a project, or all of its projects, as a downloadable archive; this requires a build environment to be running, because the archive is assembled there. An unregistered session cannot currently export, so for an unregistered session the only copy outside our systems is the one your browser holds on your device — which is lost if you clear site data. You may also connect a repository you control. We expect to make export available without a build environment, and to unregistered sessions, and will update this section when we do.

The practical consequence is worth stating plainly, because it cuts against what people expect of a hosted service: we are not a system of record for your source code. Treat hinky.dev as a place you work, and keep the authoritative copy somewhere you control.

Why the log row is a ceiling and not a list. On 6 October 2026 we found two logs nobody had accounted for: our API was recording complete requests and responses, and one function was recording the credential each caller presented. Both had been accumulating with no expiry at all, because the services that create those logs create them automatically and nothing in our own configuration described them. We fixed both, set a limit on every log in the account, and deleted what had accumulated. We state the rule as a ceiling because that is the only form of it that would have been true on 5 October: a list describes what you remembered, and a ceiling describes what you allow.

We may retain specific information longer when necessary for a documented security investigation, legal obligation, or legal claim. We restrict its use to that purpose and review continuing need at least every 90 days. Such exceptions do not justify retaining all projects indefinitely.

Request server-side deletion at privacy@hinky.dev. We verify requests proportionately and honor shorter legal deadlines where applicable. For anonymous projects, we may need a session or project identifier or another proportionate way to verify control; do not send API keys or passwords. Clearing browser data does not necessarily delete server copies. Server deletion does not remove independent copies on your device or at third-party services.

9 Security

We use reasonable administrative, technical, and organizational measures appropriate to the information and risks involved. These include encrypted transport, restricted administrative access, isolated build environments with no route to each other, resource and duration limits on every build, and access logging.

We do not scan your code for credentials, and we do not redact them. If you commit an API key, password, or other secret to a project, it is stored and processed as ordinary file content, it appears in the editing history, and it may be present in build output and operational logs. Do not put secrets in project files. Where a feature needs a credential, supply it through the mechanism provided for that purpose rather than in your source.

No system can guarantee absolute security. Build restrictions do not make submitted information unidentifiable or remove all risks. We provide incident notices as required by applicable law.

10 International processing and customer data

We operate in the United States. Your project content, its editing history, your account record and our logs are stored and processed in a single Amazon Web Services region in the United States (us-east-1), and are not replicated elsewhere. Support is provided from the United States.

We are being direct about one thing rather than vague. If you are in the European Economic Area or the United Kingdom, using the Service means your personal information is processed in the United States. We have not yet put a specific transfer mechanism in place for that, and we have not appointed a representative in the EU or UK. We are taking advice on both. We would rather say so than describe safeguards we have not implemented, and you should weigh that before entrusting us with anything sensitive. Write to privacy@hinky.dev if you want to know where this stands.

We generally determine the purposes and means of account, website, and security processing. Where we process personal information in a business customer’s project solely on that customer’s instructions, the relevant roles and terms must be addressed in a data processing agreement before that processing begins. These consumer-facing documents do not replace such an agreement.

11 Choices and rights

Depending on applicable law, you may request access, correction, deletion, a portable copy, restriction of processing, or information about recipients; object to processing; withdraw consent without affecting earlier lawful processing; or exercise applicable opt-out rights. These rights may be subject to lawful exceptions.

Send requests to privacy@hinky.dev. We verify identity or authority using information proportionate to the request, respond within applicable deadlines, and explain a denial where required. Where permitted, an authorized agent may act for you. To appeal a denial, reply to our response with “Privacy appeal”; we will review and respond within the applicable legal period. We do not unlawfully discriminate against you for exercising your rights.

You may complain to the relevant regulator, including your local European or UK data protection authority where applicable. Texas consumers may contact the Texas Attorney General. Availability of a right depends on the law’s applicability; we do not represent that every statutory regime applies to every user.

12 Younger users

The Service is for people aged 13 and older. We do not verify age and we collect no information about it — no date of birth, no age field, no verification record. The requirement is a condition of our Terms, which you accept when you start using the Service, and your acceptance is the only representation we have.

We say that plainly rather than implying a check we do not perform. hinky.dev is a tool for writing and building software; it is not directed at children and we do not market it to them. Collecting dates of birth in order to screen would mean holding more personal information about everyone, in exchange for a check that a determined person types around — which we judge to be the worse trade.

What we do instead is act when we know. If we learn that someone using the Service is under 13 — because they tell us, a parent tells us, or it becomes apparent another way — we will close the account, delete the work it holds, and stop processing their information, except anything the law requires us to keep. Write to privacy@hinky.dev; please do not send identity documents, which we neither need nor want.

Some countries set a higher age for consent-based processing, or require a parent or guardian to agree on a young person's behalf. We have not built a parental-consent process, so the Service is not available where verified consent is legally required. If that becomes a real circumstance rather than a theoretical one, we will implement the required notices and safeguards, and update this Policy before offering access.

13 Changes and contact

We will update this Policy when our practices change and identify the effective date. For material changes, we will provide appropriate notice through the Service or email where available and obtain consent where required. Continuing to use the Service does not substitute for consent where the law requires it.