> ## Documentation Index
> Fetch the complete documentation index at: https://hmm.heyhyper.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Integrations Overview — Connect Your Tools to Hyper

> Connect Gmail, Slack, Notion, GitHub, and more to Hyper. Integrations feed your shared AI memory so agents can search and surface team knowledge.

Hyper builds a shared AI memory layer by connecting to the tools your team already uses. When you link an integration, Hyper continuously syncs content from that source — emails, documents, messages, meeting notes, calendar events, and more — and makes it instantly searchable through any AI agent. You authorize each connection once through the desktop app, and Hyper handles incremental updates in the background so your knowledge stays current without any manual effort.

All integrations are managed in one place: open the Hyper desktop app, navigate to **Settings → Connections**, and connect or disconnect any source from there.

## Available Integrations

<CardGroup cols={2}>
  <Card title="Gmail" icon="envelope" href="/integrations/gmail">
    Sync important email threads and search your inbox with full Gmail query syntax — from:, subject:, has:attachment, and more.
  </Card>

  <Card title="Google Drive" icon="google-drive" href="/integrations/google-drive">
    Index Google Docs and retrieve full document content so agents can surface relevant written knowledge from your Drive.
  </Card>

  <Card title="Slack" icon="slack" href="/integrations/slack">
    Capture channel threads across one or more workspaces and search with native Slack syntax — from:, in:#channel, has:link, and more.
  </Card>

  <Card title="Notion" icon="n" href="/integrations/notion">
    Keep every Notion page indexed and searchable, with incremental updates triggered by Notion's change notifications.
  </Card>

  <Card title="GitHub" icon="github" href="/integrations/github">
    Search PRs, issues, and markdown files across multiple repos using filters like author:, state:, is:pr, and is:merged.
  </Card>

  <Card title="Google Calendar" icon="calendar" href="/integrations/calendar">
    Index upcoming and past events so agents can answer scheduling questions, find past meetings, and surface attendee context.
  </Card>

  <Card title="Granola" icon="microphone" href="/integrations/granola">
    Pull AI-structured meeting notes and verbatim speaker-tagged transcripts from Granola into your shared memory layer.
  </Card>

  <Card title="LinkedIn" icon="linkedin" href="/integrations/linkedin">
    Sync LinkedIn chats, posts, and profile data so agents can surface relationship context and conversation history.
  </Card>
</CardGroup>

## How Integrations Work

Every Hyper integration follows the same three-step lifecycle:

<Steps>
  <Step title="Authorize">
    You grant Hyper access to a source through OAuth 2.0 (or a secure native flow for LinkedIn) inside **Settings → Connections**. Hyper never stores your primary credentials — only the scoped access tokens it needs.
  </Step>

  <Step title="Sync">
    Hyper performs an initial sync to index existing content, then switches to incremental updates — using webhooks, push notifications, or polling — so new content appears in your memory layer within minutes of being created.
  </Step>

  <Step title="Search">
    Your AI agents can query any synced source using natural language or source-native search syntax. Results include direct links back to the original item so you can always verify the source.
  </Step>
</Steps>

<Note>
  Hyper only reads your data — it never writes back to connected sources. Your originals are always untouched.
</Note>

## Managing Connections

To connect, reconnect, or disconnect any integration, open the Hyper desktop app and go to **Settings → Connections**. Each connected source shows its sync status, last-updated timestamp, and a **Reconnect** button that appears automatically if a token expires.

<Tip>
  If an integration stops syncing, the most common fix is a token expiry. Head to **Settings → Connections → Reconnect** for the affected source and re-authorize — your previously synced data is preserved.
</Tip>