---
title: "Privacy Policy"
bin.md_icon: "fingerprint"
tags:
  - "bin.md policy"
---
![A woman writes beside a curious black Lab while a man respectfully looks away.](https://assets.bin.md/site-graphics/v1/code/privvy-clean-transparency.webp)




*Effective Date: October 1, 2026 · Last Updated: October 1, 2026*

This Privacy Policy describes how **bin.md** (operated by **GRiTT Information Services LLC, a Virginia limited liability company** — "**we**," "**us**," or "**our**") collects, uses, shares, and protects information when you use bin.md (the "**Service**"). We believe in keeping data collection to what is actually needed to run the platform, and in describing what we do accurately rather than in aspirational language.

For related documents, see our [[:cookie|Cookie & Storage Policy]] and [[:termos|Terms of Service]]. For our stance on free expression, see [[:freedm|Freedom]].

## Information we collect

**Account information.** When you sign up, authentication is handled by **Clerk**. Clerk collects and stores your email address, username, and — because we currently offer password sign-in — an encrypted credential. If you provide it, Clerk also stores your first name, last name, and profile image. Clerk also records session and security metadata (last-sign-in and last-active timestamps, verification attempts, lockout state) needed to operate authentication. We receive from Clerk a stable user identifier and basic profile fields so we can associate your documents and settings with your account.

**Usage data.** We record which documents you create, view, publish, and edit, along with timestamps, format (HTML/Markdown/JSON/YAML/etc.), and aggregate view counts. We use Cloudflare Analytics Engine and Cloudflare Workers Logs for operational telemetry. Workers Logs are currently retained on a 100% sample.

**IP addresses.** We do not store document-creation IP addresses in our database. As of September 5, 2026 the `doc.ip_address` column has been removed entirely and no document-creation IP is written to our database at any point. IP addresses of moderation reporters are stored only as salted SHA-256 hashes in `moderation_log.reporter_ip`; we do not retain the plaintext value. Separately, Cloudflare — the infrastructure provider that runs the Service — receives the IP address of every request as a normal part of routing traffic, and those addresses appear in the Workers Logs described under "Usage data" above. That logging is inherent to operating on Cloudflare’s network and is retained under Cloudflare’s policy for our plan.

**Billing information.** If you subscribe to a paid plan, payment processing is handled by **Clerk Billing**, which uses **Stripe** as its underlying payment processor. We do not store your credit card number or other sensitive payment data on our own infrastructure — card data is collected and stored by Stripe directly and never touches our servers. We receive plan status and identifiers from Clerk necessary to entitle your account.

**Content you publish.** Documents you paste into bin.md are stored in Cloudflare D1 and Cloudflare R2 and, unless you configure otherwise, are publicly accessible. See the AI processing section below for how content is used with AI features and moderation.

**Anonymous documents.** You may publish documents without an account. We record basic metadata for abuse prevention but do not record the creation IP address; the returned admin key is the only way to later modify or delete the document.

## AI features and content processing

bin.md uses AI in two ways. These are independent — you can opt out of one without affecting the other. Plus and Pro users can opt out from **Settings → Privacy**; Free users cannot opt out.

- **Semantic search (Plus+ can opt out).** We embed a prefix of each active document and store the embedding in **Cloudflare Vectorize** so that semantic search works. Embedding is performed via **Cloudflare Workers AI**. If you opt out, we immediately stop indexing your new content, we do not embed your search queries, and we remove your existing embeddings from Vectorize within one hour. Full-text search of your documents continues to work.
- **AI assistant and editing (Plus+ can opt out).** When you invoke the assistant or AI editing on one of your documents, the document content (up to ~32,000 tokens) is sent through **Cloudflare AI Gateway** to a large language model — currently **Anthropic (Claude)**, **OpenAI (GPT)**, or a Meta model — chosen by feature and context. If you opt out, no assistant or AI-editing calls are made on your behalf from any surface, and the assistant entry points in the app are hidden while the opt-out is in effect.

We do not use your document content to train third-party AI models. Requests routed through Cloudflare AI Gateway are logged with operational metadata only — request and response bodies are **not** retained by the Gateway. Requests are tagged with a keyed pseudonymous identifier derived from your Clerk user ID rather than the ID itself, so Gateway logs cannot be used to link activity back to your account without a secret we hold.

## Third-party data processors

The Service depends on the following processors. Each is bound by their own privacy and security terms. We do not share your data with any other parties for advertising, analytics resale, or any purpose beyond operating the Service and complying with the law.

| Processor | Purpose | Data shared |
| --- | --- | --- |
| [Clerk](https://clerk.com/privacy) | Authentication, session management, and billing | Email, username, name (if provided), password credential, profile image (if provided), session and security metadata; billing plan and entitlement state |
| [Stripe](https://stripe.com/privacy) | Payment processing for Clerk Billing (card processing, receipts, statutory billing records) | Your bin.md user identifier (in `customer.metadata`), email address, billing name and address if collected, payment method details (collected and stored by Stripe directly), transaction records |
| [Cloudflare](https://www.cloudflare.com/privacypolicy/) | Hosting, CDN, DDoS protection, D1 (database), R2 (object storage), KV (config/cache), Workers AI (embeddings), AI Gateway (assistant routing), Vectorize (semantic search), Browser Rendering (open-graph previews), Analytics Engine, Workers Logs | All Service traffic; content processed by AI features as described above |
| [Anthropic](https://www.anthropic.com/legal/privacy) | AI assistant and editing (opted-in users) | Document content (up to ~32k tokens) when the assistant is invoked; tagged with Clerk user ID |
| [OpenAI](https://openai.com/policies/privacy-policy) | AI assistant and editing (opted-in users) | Document content (up to ~32k tokens) when the assistant is invoked; tagged with Clerk user ID |
| [Svix](https://www.svix.com/privacy/) | Outgoing webhook delivery (for users who configure webhooks) | Webhook payloads and endpoint configuration; Clerk user ID as application UID |
| [Userback](https://www.userback.io/privacy-policy) | In-product feedback widget on every page; support ticket tracking | Clerk user ID, email, name, plan tier; user-agent; page URL; and screenshots or short screen recordings you submit with your feedback (which can capture on-screen document content). Email you send to `support@bin.md` is read by the operator first; if a message needs to be tracked as a support ticket, we forward it into Userback, which then receives your email address and the message content. Support email that does not become a ticket is not sent to Userback. **Privacy and data-subject requests sent to `privacy@bin.md` are never routed through Userback** — see "Your rights" below. |
| [Resend](https://resend.com/legal/privacy-policy) | Outbound transactional email | Email address; message content (notifications, receipts, moderation notices) |
| [Zoho Mail](https://www.zoho.com/privacy.html) | Inbound email hosting for all our `@bin.md` addresses | Sender address and full message content of any email you send us, including `support@bin.md`, `privacy@bin.md`, `legal@bin.md`, and `info@bin.md` |

## No data selling

We do not sell, rent, or trade your personal information to third parties for advertising, brokerage, or profiling purposes. Ever.

## No advertising or tracking cookies

We do not use advertising networks, behavioral tracking, or third-party analytics cookies. See our [[:cookie|Cookie & Storage Policy]] for the complete list of cookies and browser storage the Service uses.

## Data retention

- **Account data** is retained as long as your account is active. When you close your account from **Settings → Account**, we immediately hide your documents, revoke your API keys, OAuth authorizations, and webhook endpoints, remove other users' subscriptions to your documents, release your username, and delete the account identity at our identity provider. Your documents and files are then permanently erased within **30 days**, subject to legal preservation obligations for moderation records and billing records. **Account closure does not automatically cancel a paid subscription; cancel your subscription first from the billing area of your profile.**
- **IP addresses.** Document-creation IP addresses are not retained at all — the `doc.ip_address` column was removed on September 5, 2026. Moderation-reporter IPs (`moderation_log.reporter_ip`) are stored only as salted SHA-256 hashes and are cleared by a daily job 90 days after the report is filed. Request-level IP logging by Cloudflare is covered by the "Workers Logs" entry below.
- **Workers Logs** are retained per Cloudflare's default policy for the plan we use.
- **Billing records** are retained by Stripe (as Clerk Billing's payment processor) as required by law — typically seven years for tax and financial-record purposes.
- **Document version history.** Published document versions are retained up to a per-tier limit (default 200 snapshots per document). Deleted documents are soft-removed and purged from R2 in accordance with the deletion window.
- **Moderation records.** Reporter names, emails, and hashed IPs collected as part of moderation reports are automatically purged 90 days after the report is filed.
- **CSAM-related evidence.** If we discover CSAM, we are required by 18 U.S.C. § 2258A to preserve the content and associated metadata for at least one year and to report it to NCMEC. This preservation obligation overrides any other retention or deletion setting.

## Your rights

You have the right to:

- **Access** the personal data we hold about you.
- **Correct** inaccurate data.
- **Delete** your account and associated data at any time from **Settings → Account**.
- **Export** your documents at any time via the raw and format-specific document endpoints (`/v1/doc/:code`, `.md`, `.json`, `.yaml`, `.raw`) and, when the tooling ships, a full account export.
- **Opt out** of AI features (Plus and Pro tiers) as described in "AI features and content processing."

To exercise any of these rights, email us at [privacy@bin.md](mailto:privacy@bin.md). We will respond within 30 days.

We maintain `privacy@bin.md` as a dedicated channel that delivers directly to the operator's mailbox and is **never** forwarded into our support ticketing system. This means the personal data you submit in order to exercise a privacy right — including your email address and the contents of your request — is not shared with our support ticketing processor. Inbound email is hosted by Zoho Mail, disclosed in the processor table above.

General questions sent to `support@bin.md` are read by the operator and forwarded into Userback only when they need to be tracked as a support ticket.

### How we verify who you are

We have to be careful here, because a deletion request from the wrong person is indistinguishable from an attack on your account. Our verification is deliberately simple:

- **Send your request from the email address on the account.** For most requests, that is all we need. We match the sending address against the address Clerk holds for your account.
- **If you cannot use that address** — you lost access to it, or you are writing on someone's behalf — we will ask you to demonstrate control of the account another way, such as signing in and sending the request from within the Service, or identifying non-public details of documents you own.
- **We will not act on a deletion request we cannot tie to the account holder.** If we cannot verify you, we will tell you why rather than silently ignoring you.
- **We ask only for what the verification needs.** Information you send us to prove identity is used for that purpose and is not added to your account record.

If you are an **authorized agent** acting for someone else, send us written permission signed by that person, and we will still verify the account holder directly unless they have granted you a power of attorney.

### If we say no

We may decline a request — for example, where we cannot verify you, where the data is subject to a legal preservation obligation, or where fulfilling it would expose another person's information.

If we decline, we will tell you the specific reason. **You can appeal by replying to that message.** A human will re-review it and respond within 45 days with a written explanation. If you disagree with the outcome, you may contact your state attorney general — for Virginia residents, the [Office of the Attorney General](https://www.oag.state.va.us/).

## International users

The Service is operated from the United States and stored on Cloudflare's global infrastructure. If you are located in the European Economic Area, the United Kingdom, or another jurisdiction with a data-protection framework, please note:

- We do not currently target the EU/EEA/UK as a market and do not offer localized pricing or language for those regions.
- If GDPR or UK GDPR applies to your use of the Service, we will honor applicable data-subject requests (access, correction, deletion, portability, objection) — email [privacy@bin.md](mailto:privacy@bin.md).
- Our processors (Clerk, Stripe, Cloudflare, Anthropic, OpenAI, Svix, Userback, Resend) are US-based. International transfers rely on each processor's own Standard Contractual Clauses and safeguards.
- We do not have an EU representative (Art. 27 GDPR) at this time.

## Security

We take reasonable technical and organizational measures to protect your data. The Service runs on Cloudflare infrastructure with encryption in transit (HTTPS/TLS). Clerk handles credential storage with industry-standard security practices. Cloudflare D1, R2, KV, and Vectorize provide encryption at rest per Cloudflare's platform commitments.

No security measure is foolproof. If you discover a vulnerability, please report it responsibly to [support@bin.md](mailto:support@bin.md).

## If there is a data breach

We would rather tell you what we will do now than improvise it under pressure later.

If we determine that unauthorized access to personal data has occurred, we will:

1. **Contain first**, then investigate scope — what data, whose, and over what period.
2. **Notify affected users by email** at the address on the account, without undue delay and no later than **72 hours** after we confirm a breach affecting personal data. If we cannot yet confirm scope within 72 hours, we will send what we know by then and follow up rather than waiting for a complete picture.
3. **Tell you what actually happened** — the categories of data involved, what we know about how it occurred, what we have done, and what you should do. Not a generic advisory.
4. **Notify regulators where required.** Virginia's breach-notification statute (Va. Code § 18.2-186.6) applies to us, and where GDPR or UK GDPR applies, the 72-hour supervisory-authority deadline runs from the moment we become aware — not from when we finish investigating.
5. **Publish a post-incident summary** on the Service once the incident is closed, unless doing so would create further risk.

Because several of our processors hold data on our behalf, a breach may originate with one of them. We will notify you about a processor breach on the same terms as our own — their incident is our obligation.

**Honest limitation:** bin.md is operated by one person. We have no 24/7 security operations center, and our detection depends substantially on Cloudflare's and Clerk's monitoring rather than our own. That is a real constraint on how quickly we would notice an incident in the first place, and you should weigh it when deciding what to store here.

## Children

bin.md is not directed to children under 18, and we do not knowingly collect personal information from anyone under 18. If you believe a minor has created an account, email [privacy@bin.md](mailto:privacy@bin.md) and we will delete it.

## Known gaps and current work

We are committed to accuracy over marketing. As of the Effective Date, this policy accurately describes what our code does. Known gaps will be listed here as they arise.

## Contact

- **Privacy questions and data-subject requests:** [privacy@bin.md](mailto:privacy@bin.md) — delivered directly to the operator, not routed through support ticketing.
- **General support:** [support@bin.md](mailto:support@bin.md) — read by the operator; forwarded to Userback ticketing only when a message needs tracking.
- **Legal notices:** [legal@bin.md](mailto:legal@bin.md)

## Stay notified

:fa-bell: This document changes as bin.md does. Subscribe to this document to be notified whenever it’s updated.