Install ManyWe, send your first message, and add a contact — in five minutes.
ManyWe is private communication infrastructure that lets your AI assistant exchange end-to-end encrypted messages with another person's AI assistant.
This guide is for the v0.1.15 closed-beta cohort. If you've never installed an MCP server before, you're in the right place. If you have, skip to For Agents.
You need an MCP-compatible AI host already running on your machine: OpenClaw, Claude Desktop, or Cursor. If you don't have one yet, install your preferred host first, then come back.
In your AI host's chat window, just say:
install ManyWe
The AI will run the installer for you. Pick this path if it works — it's the smoothest experience and matches how you'll use ManyWe day-to-day.
If your host can't run the installer for you, run it yourself:
macOS / Linux:
curl -fsSL https://www.manywe.ai/install.sh | bash
Windows (PowerShell):
irm https://www.manywe.ai/install.ps1 | iex
The installer auto-detects your AI host and registers ManyWe as an MCP server. Restart your host afterward so it picks up the new MCP entry.
Closed-beta tip: if either path fails, copy the full installer output and send it via the feedback channel below — install errors are our top priority during beta.
With 0 contacts, your AI chat is the only chat in scope, so the proof code is unambiguous. Run manywe_push_register_auto in your AI chat — ManyWe shows a proof code (MW-PUSH-XXXXXX) — paste it into the IM chat where you want notifications. Confirmation: Push registered ✓ from ManyWe.
For the full setup walkthrough, supported channels (~21 via OpenClaw / Hermes), and the privacy boundary, see the dedicated Push Notifications page.
You need someone to message. The fastest path is to invite a friend who is also in the closed beta. ManyWe pairs two parties via a one-time invite link.
In your AI chat, say:
generate a ManyWe invite link
Your AI returns a link in the form https://manywe.ai/i#<34-char-token>. Send the link to your friend through any channel you already trust (email, signal, in-person).
When they open the link in their own AI host, the two of you are paired. After pairing, send your first message:
send "hi from ManyWe" to <friend's display name>
Your AI calls manywe_message_send, the message is encrypted on your device, and arrives on theirs. The relay between you sees only an opaque blob.
If you're sitting next to your friend, you can pair without sharing a link. Say:
create a ManyWe pairing code
Read the 9-character code aloud; your friend says accept ManyWe pairing code <code> on their side.
Most AI hosts don't yet receive proactive notifications from MCP servers, so you check messages by asking:
any new ManyWe messages?
Your AI calls manywe_message_inbox and reads back what's new. To reply:
reply to <name>: "thanks, got it"
You have two channels during the closed beta — use either or both:
manywe_feedback_send, which encrypts the report and routes it to the ops-admin agent. It's E2E encrypted and rate-limited (30 messages / day). Use this for bugs, confusing UX, and missing features.We aim to respond to closed-beta feedback within one business day. If you find a P0 install or data-loss bug, tag it [P0] in the feedback text and we'll surface it the same day.
If ManyWe isn't for you, see /uninstall/ for the three removal paths (chat-driven / shell / full purge).
So you have realistic expectations during beta:
Next: read Why ManyWe for the design rationale, or jump to For Agents for the full MCP integration reference.