Security at DailyDashboard

You keep client meetings, notes and invoices here, so you should know how they are protected. This page describes what DailyDashboard actually does today, in plain terms.

Sign-in

You sign in with your Google account through Firebase Authentication.

  • Your password is entered with Google, never on DailyDashboard, and we never store it.

  • There is no separate DailyDashboard password to reuse, leak or reset.

Minimal access to your Google account

DailyDashboard asks Google for the narrowest permissions that make the product work.

  • Calendar: read-only (calendar.readonly). DailyDashboard can see your events but cannot create, change or delete them.

  • Drive: drive.file only. DailyDashboard can open files it created and files you pick; it cannot browse the rest of your Drive.

  • Profile: your name and email address, to identify your account.

  • The optional GitHub integration is read-only: repository metadata, contents, pull requests and issues.

  • The optional Jira, Linear, monday.com, ClickUp and Azure DevOps integrations are read-only too: projects, boards and issues, never changes to them.

Encryption and secrets

Credentials that unlock other services are encrypted before they are stored.

  • Google OAuth tokens and the AI API keys you add are encrypted with AES-256-GCM before they are written to the database. So are GitHub, Jira, Linear, monday.com, ClickUp and Azure DevOps tokens and the Stripe keys you connect for client payments.

  • The encryption key and other server secrets (platform API keys, Stripe keys, OAuth client secrets) live in Google Cloud Secret Manager and are bound to our backend functions at runtime. They are never sent to the browser.

  • All traffic is served over HTTPS, and HTTP Strict Transport Security tells browsers to never connect over plain HTTP.

Your data stays in your account

Database security rules decide who can read every record, not just the app’s screens.

  • Clients and everything inside them (meetings, tasks, follow-ups, decisions, notes), meeting records, invoices, payments, expenses and credits can only be read by the account that owns them. The only exception is a small set of DailyDashboard administrators, for support.

  • Sensitive collections, such as stored integration credentials, connected Stripe accounts, OAuth state and webhook records, are closed to browsers entirely and only reachable by our backend.

  • Your own AI keys can be added or removed only through the backend, which validates and encrypts them; the browser cannot write them directly.

Web and API protections

Every page is served with strict security headers, and the API is rate limited.

  • A Content Security Policy that only allows scripts from DailyDashboard, Google and Stripe; the few inline scripts must match an allow-listed hash.

  • X-Frame-Options DENY, so DailyDashboard cannot be embedded in another site to trick you into clicking.

  • Strict-Transport-Security with preload, X-Content-Type-Options nosniff, a strict Referrer-Policy and a Permissions-Policy that turns off the camera, location and other sensors.

  • Rate limits on the API, with separate, tighter limits for admin operations and per-user limits for the Claude connector.

Payments

Card details are handled by Stripe and never touch DailyDashboard’s servers.

  • Subscriptions are bought through Stripe’s pricing table and checkout, and managed in Stripe’s customer portal.

  • Stripe webhooks are verified with their signing secret before we act on them.

  • If you connect your own Stripe account to record client payments, a restricted key with read access to charges and customers is enough. The key and webhook signing secret are encrypted, and only payments made after you connect are recorded.

Claude connector (MCP)

Connecting Claude, or another MCP client, to your account uses OAuth 2.1 and is always your decision.

  • PKCE (S256) is mandatory on every authorization; the weaker "plain" method is rejected.

  • You approve access on a DailyDashboard consent page that shows which app is asking, the host it will return you to, and the scopes it wants.

  • Access is split into read and write scopes. Write tools (creating or updating tasks, follow-ups, decisions and client details) only work on connections approved with the write scope.

  • Access tokens expire after 1 hour. Refresh tokens rotate on every use, and reusing an old one revokes the whole connection.

  • Tokens are opaque random values. DailyDashboard stores only their SHA-256 hashes, and connector tokens are accepted by the connector endpoint and nothing else.

  • Every request re-checks that your account still has agent access.

  • Revoke any connection at any time from Settings → Integrations → AI assistants → Connected apps. It takes effect immediately.

AI providers and your keys

You choose which AI provider runs your features, and whose key pays for it.

  • Pick Claude (Anthropic), OpenAI or Gemini (Google) in Settings → Integrations → AI Provider, or leave it on Automatic. You can also choose models per feature.

  • Bring your own API key for any of the three. It is checked with the provider before it is saved, stored encrypted, used instead of the platform key, and removable at any time.

  • Voice transcription, text-to-speech, embeddings and knowledge-base search always run on OpenAI, using your OpenAI key if you have added one.

Agents act within limits you set

Agents only use the tools you give them.

  • Choose each agent’s tools, and whether it works across all clients or only the ones you pick.

  • Run an agent in manual mode and it proposes actions, then waits for your approval before doing anything.

  • New agents hold emails from unknown senders for your approval, and hold their email replies for your review, by default.

What this page does not say

This page lists only protections that exist in the product today. It does not describe certifications, audits or penetration tests, and it makes no claims about how AI providers retain or use the data they process: that is set by each provider’s own terms. If that matters for your clients, add your own API key so requests run under your own account with that provider. Our Privacy Policy covers what we collect and why.

Report a security issue

If you think you have found a vulnerability, please email us with the details and steps to reproduce it. Please do not access other people’s data or disrupt the service while testing. We will follow up with you by email.

support@dailydashboard.ai

Using the Claude connector? See how to connect Claude to your practice.