Privacy / Current practices
How your data is handled
Updated
This page describes how Proper Respect handles accounts, evidence and publication as of October 6, 2026.
Proper Respect is created by lecturesfrom, a business in the United States, and operated by Keegan Moody. Contact [email protected] with questions about your data.
Your account and private collection
Clerk handles sign-in. Convex stores application records and uploaded evidence. Account records include an authentication identifier, your handle, display name, profile information and saved links. Your collection can contain product relationships, notes, dates, costs and supporting evidence.
Private collection access is tied to your signed-in identity. Evidence records can include source URLs, original uploads, message metadata, observations, excerpts and capture history. These records are stored by the service; they are not kept only on your device.
Aggregate request tally
When enabled, the footer tally counts eligible page requests on public pages and on private, sign-in, sign-up and onboarding pages. It stores only aggregate category totals and a start timestamp in Upstash Redis. Visitor IP addresses, account identifiers, request paths or raw User-Agent strings are not sent to this store by the tally.
Categories are inferred from what a client claims in its User-Agent header, and those claims can be false. Requests with missing identification are counted separately as unidentified. Totals measure requests, not unique people or use of a listed product. An AI-assistant or link-preview category does not establish that a person requested a page, shared a link or endorsed a product.
The tally excludes static assets, API requests, prefetches, prerenders and client-side data fetches. Hits can be dropped when the store or request background processing is unavailable. Displayed totals can be delayed or incomplete; they are not a complete traffic record.
Connected sources
Connecting Gmail grants read-only Gmail access, a broader permission than the metadata the current reader requests. The reader requests message identifiers, timestamps and the From, Subject and Date headers. It does not request message bodies or attachments. These headers can contain personal information, and unmatched headers can be retained as private evidence.
New and reconnected mailboxes start with scheduled maintenance disabled. The collection provides controls for supported reads and maintenance. Saved mailbox credentials are encrypted by the application; this does not mean all stored evidence is end-to-end encrypted.
Supported activity connections can retain snapshots and their measurement periods. For example, the GitHub connection requests account information and a contribution calendar for a one-year period. A snapshot does not establish complete usage, every repository or all work performed.
Proper Respect's use and transfer to any other app of information received from Google APIs will adhere to the Google API Services User Data Policy, including the Limited Use requirements.
What becomes public
Saving a collection and publishing it are separate actions. You select saved cards and details, review an exact preview, then approve publication. Visitors and browser agents can read the resulting public profile without signing in. Raw mailbox evidence and credentials are not part of that public profile.
Review your display name, biography and links as well as the cards in the preview. A display name may initially come from your sign-in identity, including an email address. Removing every public card can leave your public name, profile information and an empty collection visible; it does not delete the published identity.
Disconnecting and deleting are different
Disconnecting a mailbox removes its saved credentials and stops future reads through that connection. Existing private evidence, source history and relationship history are retained. Disconnecting a Gmail account also asks Google to revoke this app's access. If that request fails, the local disconnect still completes, and you can revoke the app yourself in your Google Account permissions.
Activity-connector disconnection removes the local saved secret and stops that connection's updates. Existing snapshots, relationships and published cards can remain. It does not establish that a linked provider account or its grant has been removed.
Deleting an original evidence payload does not erase all information derived from it. Observations, excerpts, source identifiers, hashes, provenance, relationships and published records may remain. A complete self-service account deletion workflow and automatic evidence-expiry schedule are not implemented. Disconnecting or deleting an original does not fully erase your data.
Services and browser storage
Product presentation can use Context.dev and official product sources. Brand lookups send the product domain to the branding service. Product images and fonts can load from external hosts, so visiting a page can make requests to those hosts.
Your appearance preference is saved in your browser's local storage. Authentication and hosting services also process requests and may use cookies and logs as part of their services.
Searchable Analytics counts visits to public pages on our production website in an isolated browser frame. It receives public paths without query strings or fragments, plus browser and device signals. The frame receives no page content, page title or referrer from the parent page. Signed-in, sign-in, sign-up and onboarding routes are excluded. Cookies and interaction plugins are disabled, and the frame cannot access parent-page cookies or browser storage. Its temporary visitor and session identifiers reset after a full page load or a visit to an excluded route. This browser integration does not measure referrals from AI services.
Proper Respect uses PostHog for product analytics. On public pages it records page views, clicks, page performance, JavaScript errors and session replay, with every form field masked. On signed-in, sign-in, sign-up and onboarding pages it sends no replay, clicks or errors: only a page view reduced to the route name, such as /app, without card names, IDs or page text. Visitors are not identified to PostHog by account. Requests from bots and crawlers are labeled as bots rather than dropped, so human and automated visits can be counted separately. PostHog stores this data in the United States and uses cookies and local storage to recognize a returning browser.
How to request deletion
Email [email protected] from the address you sign in with. Say whether you want your whole account deleted, specific evidence removed, or your public profile taken down, and include your handle. Do not send passwords, tokens or private originals.
Deletion is handled by hand while a self-service workflow is not available. We will reply to confirm what was deleted and to name anything that remains, such as records described in the section on disconnecting and deleting above.
Questions about your data
Contact Keegan Moody at [email protected] about access, retained evidence or deletion. Describe what you need without sending passwords, tokens or private originals. The disconnecting and deletion section above explains what the application currently retains.