For Bee AI Wearable Owners
Your transcripts,
organized.
Finally a home for everything Bee captured.
Bee records your day — but browsing 3,003 conversations in the app is rough.
MemoryHive gives you instant search, fact clustering,
and a one-click Obsidian vault export.
Zero cloud. Your data stays on your machine.
I built this for myself after syncing 3,000+ conversations and having no good way to search them. Now I'm sharing it. — Brian
Help Shape the Future of Personal Wearable AI
Be among the first to organize, search, and secure your Bee AI wearable transcripts. We are onboarding early testers and developers for exclusive direct support and full access.
Join the Alpha Testing Group
We are onboarding alpha testers on a rolling basis. Get early dev builds, direct support, and sync your transcripts.
Source is shared under BUSL-1.1 — view, fork, self-host for personal use. No commercial redistribution. Converts to MIT in 3 years. Learn more ↓
Explore the app — click any section
100% Local & Private
Your transcript data never leaves your machine. SQLite database, offline processing, zero telemetry. No accounts, no servers.
Instant Full-Text Search
Sub-millisecond search across 3,000+ conversations. Filter by date range, person, tag, and location simultaneously.
Obsidian Vault Export
1-click export builds a structured vault: daily notes, weekly summaries, people pages, tags, and wikilinks — ready to open.
Smart Fact Management
Review, bulk-edit, and cluster every fact your Bee extracted. Adjustable similarity threshold, keyboard-first workflow.
Choose your edition
Desktop App
💡 Pricing Survey: We are trying to figure out our pricing strategy. Please click on the price you would like to pay for such a service!
- Zero-setup double-click install (macOS & Windows)
- Background server launcher — no terminal needed
- Auto-updates included
- All features: search, export, facts, journals, todos
- Direct Bee account sync built in
- Supports ongoing development
↑ Looking for alpha users — sign up above
💡 Pricing Survey: We are trying to figure out our pricing strategy. Please click on the price you would like to pay for such a service!
Self-Hosted Edition — run it on your own server
- Docker Compose — one command setup
- GitHub source + auto-update script
- All app features
- 1 user account
- Multi-user support
- Docker Compose — one command setup
- GitHub source + auto-update script
- All app features
- Up to 5 user accounts
- Shared server for the whole household
- Docker Compose — one command setup
- GitHub source + auto-update script
- All app features
- Unlimited user accounts
- Every future release, no renewal
Built in the open.
Protected to stay that way.
MemoryHive's source code is available on GitHub for developers to read, learn from, and self-host. The codebase is under the Business Source License 1.1 — the same model used by Sentry, HashiCorp, and MariaDB.
This keeps the code open for inspection and personal use while protecting against someone forking it and selling a competing product. The license converts to MIT after 3 years.
Request GitHub Access →What BUSL-1.1 means for you
Your memory
deserves better tools.
Get Early Access
We'll notify you the exact second local desktop builds are ready for download.
View, fork, and self-host. No commercial redistribution. Full license terms ↑
Or support the project —