Welcome to the Preview

Thanks for joining the MemNexus gated preview. You're one of a small group of developers testing the full product before public launch. This guide walks you through everything — account setup to your first memory to exploring the dashboard.

What you get

• Full product access — CLI, SDK, REST API, dashboard, and billing
• Free tier — 200 memories, 60-day retention
• Pro tier — Unlimited memories, unlimited retention (simulated billing via Stripe test mode)
• Direct line to the team — Discord server for real-time chat, GitHub Issues for bug reports

Get set up

cmk_live_abc123.sk_your_secret_here

npm install -g @memnexus-ai/cli

mx --version

mx auth login --interactive

mx auth status

Create your first memory

Now the fun part. Create a memory from the CLI:

mx memories create \
  --content "I'm testing MemNexus as part of the gated preview. First impression: setup was fast." \
  --topics "preview,first-impression"

You should see a confirmation with the memory ID.

Search for it

mx memories search --query "what was my first impression of memnexus"

The semantic search finds your memory by meaning, not exact keywords. Try different queries — "preview setup experience" or "how did onboarding go" — and see how the search handles them.

View it in the dashboard

Go to your dashboard and navigate to Memories. You'll see the memory you just created, with its content, topics, and metadata. You can edit or delete it from here.

Use it with your AI coding assistant

The CLI works with any command-line AI assistant — Claude Code, GitHub Copilot, Cursor, or anything that can run shell commands. No MCP setup needed. Just tell your assistant to use mx commands.

Save a memory

Tell your assistant something like:

You: "Use mx to save a memory that our API rate limit is 100 requests per minute per key, and we chose that based on the P95 usage from last month's analytics."

Your assistant runs:

mx memories create \
  --content "API rate limit is 100 req/min per key. Decision based on P95 usage data from last month's analytics." \
  --topics "api,rate-limiting,decision"

Retrieve it in a new conversation

Start a fresh conversation with your assistant and ask:

You: "Use mx to look up what our API rate limit is and why we picked that number."

Your assistant runs:

mx memories search --query "API rate limit decision"

And gets back your memory with the full context — even though this is a brand new conversation.

That's the core idea. Knowledge persists across conversations and across tools. What Claude Code learns today, Cursor can recall tomorrow.

Explore billing

During the preview, billing runs in Stripe test mode — no real charges. To try the upgrade flow:

1. Go to Billing in your dashboard
2. Click Upgrade to Pro
3. Use Stripe's test card: 4242 4242 4242 4242 (any future expiry, any CVC)

This lets you experience the full Pro tier: unlimited memories and retention.

What's next

Something broken? Have an idea? Let us know. Join the Discord for real-time help.