AI MEMORY LAYER · MCP-NATIVE

Your notes are
your AI's memory.

Collections you and your AI both edit. One memory, synced across every device and every AI.

Get early access See how it works

en·gram: a unit of cognitive information imprinted in a physical substance.

HOW IT WORKS

Three steps.
No more re-explaining.

  1. Step 01

    Bring your notes.

    Files on your computer that you and your AI both read natively, with no proprietary format. Already use Obsidian? Point Engram at your existing vault. Starting fresh? Open the web portal and write.

  2. Step 02

    Connect to any AI.

    Engram speaks MCP, the open standard Claude, Cursor, and ChatGPT use to reach external tools. Plug in once and your AI can read your notes, write new ones, and update existing ones.

  3. Step 03

    Ask anything.

    Your AI pulls the right notes itself. No more pasting context. No more "what did we decide last quarter?" It already knows.

INSTANT CONTEXT

Ask.
It already knows.

Skip the re-explanation. Your AI pulls from notes you already wrote and answers like it remembers. Because, through your notes, it does.

You: "What did we decide about the kitchen renovation last quarter?"

Claude: Pulling from /home/2026-Q1/kitchen-decisions.md and 3 linked notes…

"You chose slate countertops over quartz because of the light-and-shadow contrast you noticed at the showroom. Budget came in 12% under thanks to the tile reuse plan."

Compatible with

  • CLAUDE DESKTOP
  • CURSOR
  • CHATGPT DESKTOP
  • CLAUDE CODE
  • CLINE
  • ANY MCP CLIENT
Read the full MCP guide
DIFFERENTIATOR

One vault.
Edited from anywhere.

A vault is a folder of markdown notes you own. Engram syncs it between every device you write on: desktop, mobile, tablet, browser. And while it's syncing, it indexes every word so any AI you connect can recall it.

  • In Obsidian, on every device.

    Desktop, mobile, tablet: your vault stays in sync without iCloud hacks or a separate Obsidian Sync subscription. Same notes everywhere you open them.

  • In the browser.

    app.engram.page: open and edit on any machine.

  • By your AI.

    Claude, Cursor, ChatGPT: every compatible AI reads and writes.

  • Never lose an edit.

    Offline edits queue and replay on reconnect, deduplicated by path.

OBSIDIAN WEB # VAULT CLAUDE CURSOR CHATGPT
  • OBSIDIAN
  • WEB PORTAL
  • MCP
  • BIDIRECTIONAL
WHO IT'S FOR

Anyone tired of
repeating themselves.

  • Researchers & writers

    Years of reading notes, source quotes, and half-finished arguments, all instantly retrievable. Ask a question and your AI cites your own marginalia, not Wikipedia.

  • Engineers & PMs

    Architecture decisions, postmortems, RFCs. Stop pasting "here's the context" into every Cursor session. Your AI already knows the system.

  • Founders & operators

    Customer calls, board prep, pricing experiments. Your AI becomes a chief of staff that has read every note, not a stranger you re-onboard daily.

  • Personal Knowledge Management enthusiasts

    Already running Obsidian with thousands of notes? Engram makes them queryable by every AI you use, without leaving the editor you love.

PRIVACY

You own the keys.
You own the exit.

Encrypted at every step of the journey.

Encrypted in transit. AES-256-GCM at rest with per-user keys: notes, attachments, paths, filenames, every byte. The server searches your vault through HMAC fingerprints, never plaintext.

Coming

Bring your own keys.

Hook AWS KMS as the master-key root and Engram won't hold the key that unlocks your data. Revoke access in KMS and your vault goes dark instantly, even to us. Shipping after launch.

Never sold. Never trained on.

Your notes don't feed anyone's model, including ours. No analytics resale, no third-party data brokers.

Plain markdown. Plain folders.

No proprietary format. Cancel any time and walk away with every note in a folder you can read in any text editor.

  • AES-256-GCM AT REST
  • HMAC BLIND INDEX
  • NO AI TRAINING
  • BYOK · AWS KMS
  • NO TELEMETRY
SELF-HOSTABLE

Skip the cloud entirely.
Run it yourself.

Server, sync engine, web portal, embeddings: every byte runs on hardware you control. The strongest privacy posture we can offer, available out of the box. Rip the cord on us any time and keep going.

Docker Compose ready.

One command brings up the whole stack. No Kubernetes degree required. Runs on a homelab box, a NAS, a $5 VPS, or whatever's already humming in the closet.

Source-available. Read every line.

Server, plugin, and web portal are public on GitHub under the PolyForm Small Business License 1.0.0, with a separate commercial license for larger organizations. Personal use and small-business self-hosting are unrestricted. Audit the crypto, fork the sync engine, send a PR.

Local AI for maximum privacy.

Pair Engram with Ollama or any OpenAI-compatible local model and your vault never leaves the LAN. Pay nobody for embeddings. The only network call your notes make is the one between two of your own machines.

No phone-home, no license server.

Cut the WAN cable and the stack keeps running. No activation handshake, no usage beacons, no kill switch we can flip. Air-gap it if you want.

  • DOCKER COMPOSE
  • SOURCE-AVAILABLE · POLYFORM SBL
  • LOCAL EMBEDDINGS
  • OLLAMA COMPATIBLE
  • AIR-GAP READY
View the source on GitHub
PRICING

Three tiers.
Free to start. Pay when you scale.

Pay monthly or annual. Annual saves ~17%.

  • Free

    $0

    • 1 vault, 1 device
    • 10k notes · 1 GB
    • Manual sync
    • MCP integration
    • 5 AI conversations/day

  • Starter

    $7/mo

    or $70/year

    • 5 vaults, unlimited devices
    • 50k notes · 3 GB
    • Real-time sync
    • 500 AI queries/day
    • Full API + MCP
  • RECOMMENDED

    Pro

    $14/mo

    or $140/year

    • 15 vaults · unlimited notes · 15 GB
    • Unlimited AI (fair use)
    • Smart retrieval (coming)
    • BYOK / custom model (coming)
    • Priority support
See full comparison
FAQ

Common questions.
Quick answers.

  • Do I need Obsidian?

    No. The web portal at app.engram.page is a full editor. Obsidian is great if you want a desktop-native experience. Install the plugin and your existing vault syncs.

  • Does it work with my AI tool?

    If it speaks MCP, yes: Claude, Cursor, ChatGPT Desktop, and a growing list of clients. The goal is to work everywhere your AI lives. Don't see yours? Email hello@engram.page with the connection you want and we'll line it up.

  • Can I import my existing notes?

    Yes. Drop a folder of markdown files into the web portal or point the Obsidian plugin at an existing vault. Engram indexes in the background while you keep working.

  • What if my AI writes the wrong note?

    Every AI write is logged with timestamp and source. The web portal shows per-note history with one-click revert. You're never locked into an AI's mistake.

  • What about offline?

    Edits queue locally and replay on reconnect, deduplicated by note path. The Obsidian plugin works offline by default.

  • Mobile?

    The Obsidian plugin runs on iOS and Android. The web portal works in any modern mobile browser.

  • How do conflicts get resolved?

    Path-deduplicated, last-write-wins per note, with per-note edit history so nothing is silently lost. Concurrent edits to different notes simply merge.

EARLY ACCESS

Stop re-explaining.
Start remembering.

Join the waitlist and we'll email you a one-click checkout link when Engram is open.

By joining, you agree to our Privacy Notice. We collect your email to send launch updates. We don't sell or share personal information.

One confirmation email, then nothing until launch. Unsubscribe any time.