Mirador Privacy Policy
Effective date: July 29, 2026
Mirador ("Mirador," "we," "us") is a calendar and scheduling service operated by [ENTITY]. It unifies your connected calendars, mirrors events between them, hosts booking pages, and provides an AI assistant ("Mira"). This policy explains what we collect, how we use it, and the controls you have. Contact for anything in this policy: pietro@gnarlysoft.com.
1. Information we collect
Account information. Name, email address, optional avatar, time zone, and sign-in credentials (a password hash, magic-link tokens, passkey public keys, or single sign-on identifiers — we never see your identity provider password). Session records include IP address and browser user-agent.
Calendar data. When you connect a calendar (Google or Microsoft), we store the events needed to operate the service: title, start/end time, location, a link back to the event in your provider, and sync metadata. We store the number of attendees on an event, not their names or addresses. We store the OAuth tokens you grant so we can sync; the scopes are shown at connection time.
Email data. For connected mailboxes, we sync a rolling window of recent messages and store metadata and a short snippet only: sender/recipient name and address, subject, an up-to-200-character preview, folder, timestamps, and read state. We do not store email bodies — when a feature needs the full text (for example, drafting a reply you asked for), it is fetched live from your provider, used, and not retained.
AI writing profile. With your permission (on by default, with an off switch in Settings → Data & privacy), Mira periodically derives a short summary of your writing style — tone, greeting, sign-off — from your own sent messages. Only this summary is stored, not the messages. You can turn this off or clear the stored profile at any time.
Booking data. When someone books time through a public booking page, we collect the name, email address, and answers they submit, and share them with the host (including adding them to the host's calendar event).
Assistant and product data. Your chats with Mira, actions the assistant takes (with an activity log), AI-generated meeting briefs (which include attendee names/emails for that meeting), notifications, support tickets you file, and API (MCP) keys you create (stored hashed; tool calls are logged by name, not content).
Audit log. Security-relevant events — sign-ins and sign-outs (with method), session revocations, SSO configuration changes, data exports and deletions, and administrative actions — recorded with actor, timestamp, and action.
What we don't collect: no advertising identifiers, no analytics or tracking scripts, no payment card data (paid billing is not yet live; if introduced, payments will be handled by a payment processor and card numbers will not touch our servers).
2. How we use information
To provide the service: syncing and mirroring calendars, computing availability, operating booking pages, sending the transactional email the service requires (sign-in links, booking confirmations and changes, team invitations, and the optional morning brief), providing the AI features you use, securing accounts, and offering support. We do not use your information for advertising, and we do not sell or share personal information for advertising purposes.
3. AI processing
Mirador's AI features run on Anthropic Claude models via Amazon Web Services (Amazon Bedrock), and help-center search uses Google Gemini embeddings. What is sent, per feature:
| Feature | Content processed | When |
|---|---|---|
| Mira chat | Your messages, plus workspace context (calendar labels/account emails, a week of event titles/times) | When you chat |
| Email triage | Sender, subject, and snippet of recent messages | Background |
| Meeting briefs | Meeting title, attendees, related email subjects/snippets, booking answers | Background, before meetings |
| Morning brief | Your day's events and a few email subjects | Background, daily (if enabled) |
| Draft replies | The email you're replying to (fetched live) and your writing profile | When you ask for a draft |
| Writing profile | Your own sent messages (unless opted out) | Background, periodic |
| Help assistant | Your typed question | When you ask |
Model providers process this content to return a response; your data is not used to train AI models, and we have no model-training integrations. AI processing happens in United States AWS regions. AI output can be inaccurate — features that act (like sending an email) always require your confirmation.
4. Sharing and service providers
We share personal information only with the processors that run the service, under their respective data-processing terms:
- Amazon Web Services (US) — hosting, database, email delivery (SES), and AI inference (Bedrock; may process within multiple US regions).
- Google — if you connect Google Calendar (and optionally Gmail read-only), data flows to/from your Google account under the scopes you grant; Gemini processes help-center questions; our pages load fonts from Google Fonts, which receives your IP address and user agent.
- Microsoft — if you sign in with Microsoft or connect a Microsoft calendar/mailbox, data flows to/from your Microsoft account under the scopes you grant.
- Booking hosts and workspace members — information you submit when booking is shared with the host; content in a shared workspace is visible to its members per their roles.
We may also disclose information if required by law, or in connection with a merger or acquisition (with notice).
Google API Limited Use disclosure. Mirador's use and transfer of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements. Google user data is used only to provide the user-facing features described here, is never used for advertising, and is never read by humans except with your explicit permission, for security purposes, or to comply with law.
5. Cookies
Mirador sets only cookies necessary to operate the service: the session cookie and short-lived authentication-flow cookies (all httpOnly), a cached session-profile cookie (5-minute lifetime), and a cookie remembering your active workspace. There are no analytics, advertising, or cross-site tracking cookies.
6. Public pages
Booking pages you enable are public web pages. They display the workspace name and logo, host names, titles, and avatars, meeting types, and available time slots. Availability is computed from hosts' calendars, but visitors only ever see free/busy — never event titles, attendees, or details. If you don't want this information public, disable the booking link.
7. Your rights and controls
Self-serve, in the app:
- Export your data (Settings → Data & privacy) — a JSON file of your profile, memberships, chat history, and (for workspaces you own) calendar, email metadata, booking, and audit data.
- Delete your account (Settings → Data & privacy) — permanently removes your account, connections, and workspaces only you own. Where we attempt it (Google), we also revoke our access tokens at the provider; you can additionally revoke Mirador's access anytime from your Google or Microsoft account settings.
- Delete a workspace (Settings → Workspace → General; owners).
- Disconnect a calendar or mailbox (Settings → Connected accounts) — deletes its synced data and our tokens.
- AI writing-profile opt-out and clear (Settings → Data & privacy).
- Revoke sessions and revoke API keys (Settings → Security; Connections).
For anything else — including access, correction, or deletion requests you can't complete in-app — email pietro@gnarlysoft.com. We respond to verified requests within the timelines applicable law requires.
California residents (CCPA/CPRA): you have the right to know, access, correct, and delete personal information, and to non-discrimination for exercising those rights. We do not sell or share personal information as defined by the CCPA, and we do not use sensitive personal information beyond providing the service. Submit requests via the email above; an authorized agent may act for you with proof of authorization.
EEA/UK and Brazil: where GDPR or LGPD applies, we process personal data to perform our contract with you, for legitimate interests in operating and securing the service, or with your consent, and you have the corresponding statutory rights (access, rectification, erasure, portability, objection, and complaint to your supervisory authority).
8. Retention
We keep personal data while your account is active. Deleting your account, a workspace, or a connection removes the associated data from our production database as described above. Audit logs may be retained longer for security and compliance purposes. Booking records emailed to participants remain in the participants' own mailboxes and calendars.
9. Security
All traffic is encrypted in transit (TLS). Access to production systems uses role-based cloud IAM (no shared credentials); the application enforces role-based access control, workspace isolation, and an append-only audit log. Enterprise workspaces can require SAML single sign-on with proof of domain ownership. Our staff support tools expose account, plan, and support-conversation data — they do not provide access to customers' email content, calendar event content, or assistant chats. No method of storage or transmission is perfectly secure; report suspected vulnerabilities to pietro@gnarlysoft.com.
10. Data location
Mirador runs in Amazon Web Services in the United States (primary region us-east-1). If you use Mirador from outside the US, your information is transferred to and processed in the US.
11. Children
Mirador is a productivity service not directed at children and may not be used by anyone under 16.
12. Changes
We'll post changes here and update the effective date; material changes will be announced in the product or by email. Continued use after a change takes effect constitutes acceptance.
13. Contact
[ENTITY] · [ADDRESS] · pietro@gnarlysoft.com