Privacy Policy

Last updated: 2026-10-18

Who we are

ScreenBlurt (“ScreenBlurt,” “we,” “us,” or “our”) is operated by Justin Jordan, an individual doing business as ScreenBlurt at screenblurt.com.

This Privacy Policy explains what information we collect, how we use it, and your choices. It applies to the ScreenBlurt website and the ScreenBlurt macOS app.

Privacy / rights contact: support@screenblurt.com

What ScreenBlurt does with your screen content

This is important for how the product works:

• Screenshots stay on your Mac. Screen capture happens on your device. Our servers never receive screenshots and do not run OCR.
• OCR runs locally in the Mac app.
• Only text reaches our servers. The Mac app sends text for speech synthesis: text recognized in a Screenshot, text you highlighted as a Selection, or text an AI Agent you connected asks it to speak (see below). Our servers accept text, not images, and do not check where the text came from: whatever text is submitted is sent to our speech provider (OpenAI) to generate audio.

AI agents you connect (MCP)

The Mac app includes an optional Model Context Protocol (MCP) server that lets AI agents you choose to connect (for example, Claude Code, Codex, Claude Desktop, Cursor, Windsurf, or VS Code) ask ScreenBlurt to speak text to you. How it works:

• It runs only on your Mac. The MCP server is part of the ScreenBlurt app and talks to agents on your own computer. We do not operate a hosted MCP endpoint, and agents never connect to our servers directly.
• You decide which agents can use it. The first time an agent connects, the app asks you to allow or deny it. The app stores those choices (the name each agent reports for itself, and whether you allowed it) on your Mac; you can turn an agent off or forget it in Settings → Agents.
• Agent text is handled like any other text. When an allowed agent calls speak, the app sends that text through the same speech-synthesis path described above, under your signed-in account. speak sends text only; we do not collect microphone audio or a recording of your voice. We do not store the text, we do not keep a history of what agents said, we do not store the generated audio file, and our servers do not record which agent (if any) made a request.
• What the app tells a connected agent. An allowed agent can call status, which returns service information: whether you are signed in, your remaining character balance, the available voices and quality levels, your default voice and quality, the character cap, the server address the app uses, and the most recent error message (if any) from loading your account. speak results also report your remaining balance. In current versions of the app, none of this includes your email address, password, or billing information; versions before October 2026 included the account email, so update the app if you are on an older one. The agent and its provider may then process that information under their own terms.
• The agent is not ScreenBlurt. Agents and the companies that run them (for example, Anthropic, OpenAI, Cursor, or Microsoft) are not our sub-processors. What an agent sees, writes, sends to its own provider, or stores is governed by that agent’s terms and privacy policy, not this one. Our setup instructions (such as screenblurt.com/agents.md) may ask your agent to add ScreenBlurt to its configuration and save an optional /blurt command file in its settings folder on your Mac; the agent makes those changes, not ScreenBlurt.

What we do not keep (read text / content)

ScreenBlurt does not store the text you have read aloud. It is carried in a signed synthesis token that expires after about 30 seconds, and our usage and failure records keep only metadata (see below), never the text. We also do not store the generated audio file. Our hosting provider keeps standard, short-lived request logs to operate the service; those are separate from our own records, and we do not use them to read or archive your text.

Our speech-synthesis provider (OpenAI) receives the text to generate audio and may retain it for a limited period under its own API data policies; under the API terms we use, it is not used to train their models. That retention is separate from ScreenBlurt’s own 30-second token window.

What we do store

We store information needed to run accounts, billing, and the service, including:

We may also store basic operational logs (for example, errors or request metadata) needed to keep the service running and secure. Those logs are not used as a long-term archive of your read-aloud content.

How we use information

We use the information above to:

• Provide, operate, and improve ScreenBlurt (accounts, synthesis, credits, subscriptions)
• Process payments and prevent fraud or abuse
• Communicate with you about your account, support, and service changes
• Comply with law and enforce our Terms

Justin Jordan, who operates ScreenBlurt, can see, through an admin panel, your email address, Stripe billing identifiers (customer and subscription IDs), plan and credit balances, and payment, refund, and dispute history, and uses them to run the service and handle support and refunds.

Sharing and sub-processors

We do not sell your personal data.

We do share information with service providers (“sub-processors”) who help us run ScreenBlurt, only as needed for those services. Current sub-processors include:

These providers process data under their own terms and privacy policies, as applicable to their role. We may also disclose information if required by law, to protect rights and safety, or in connection with a business transfer (for example, if ScreenBlurt’s operations are sold or reorganized)—in which case we would expect the recipient to honor comparable privacy commitments where required.

We do not claim that we “never share any information.” Sharing with the providers above (and as otherwise described here) is necessary to operate the product.

Cookies and similar technologies

Cookies (and similar session technologies) are used for session purposes only — for example, keeping you signed in. We do not use advertising trackers or tracking pixels on the website or in the app.

The website uses Vercel Web Analytics to count page views in aggregate and to record a small set of custom funnel events. Page views set no cookies and do not identify you across sites or across days: they record the page visited, the referring page, your browser, operating system and device type, and an approximate location derived from your IP address (country, region and city). Credentials in page addresses, such as password-reset links, are removed before anything is sent.

Custom events record only step names and coarse details: for example that someone started sign-up, verified an email, started a trial, started or completed checkout (including the plan name: basic, plus, or pro), linked a device, or downloaded the app, and coarse step outcomes such as started or completed. No personal identifiers — email address, user ID, or name — are attached to these events. They are processed by Vercel Analytics under the same cookieless, aggregate rules as page views.

The app sends no analytics. If that changes, we will update this policy.

Your choices and rights

Depending on where you live, you may have rights to access, correct, export, or delete personal information we hold about you, or to object to or restrict certain processing.

Self-serve account deletion is not yet available. To request deletion or an export of your data, or other privacy-related help, email support@screenblurt.com. We aim to respond to valid requests within about 30 days, and we will work with you to fulfill them as required by applicable law and as our systems allow.

You can also update account email or cancel a subscription through normal product / billing flows where available, or by contacting support.

Children

ScreenBlurt is a general-audience service and is not directed at children under 13. We do not knowingly collect personal information from children under 13. Accounts may only be created by people 13 or older (or the minimum age required in your jurisdiction), and every account holder confirms this at sign-up. We do not collect a separate parental consent or send parental notices; we rely on that sign-up confirmation and this policy.

If a parent or legal guardian allows a child under 13 to use ScreenBlurt under the parent or legal guardian’s own account, the account holder is responsible for and authorizes that use, including sending any text the child has read aloud (or that an AI agent connected to the account asks ScreenBlurt to speak) to our speech provider (OpenAI) to generate audio, as described above. All account, email, and billing information belongs to the account holder. If you believe a child under 13 has created an account or provided us information directly, contact support@screenblurt.com and we will delete it.

Data security and retention

We take reasonable steps to protect account and service data. No method of transmission or storage is 100% secure.

We retain account, billing, and usage metadata for as long as your account is active and as needed for billing, support, security, and legal obligations. Read-aloud content is not stored long-term on ScreenBlurt servers, and we do not store the generated audio file (see above). After account deletion requests, we will delete or de-identify personal data we control, subject to backups and legal retention needs.

International users

ScreenBlurt is operated from the United States. If you use the service from elsewhere, your information may be processed in the U.S. and in countries where our sub-processors operate.

Changes to this policy

We may update this Privacy Policy from time to time. The “Last updated” date at the top will change when we do. We encourage you to review this page periodically. Continued use of ScreenBlurt after an update means you acknowledge the revised policy.

Contact

For privacy questions, deletion/export requests, or rights inquiries:

support@screenblurt.com

Related: Terms of Service · Refund Policy