Skip to content

Use cases

Made for your life. And your work.

One memory that every AI can use. Here’s what that looks like.

For everyday life.

Everything you’d otherwise have to remember, explain again, or dig out of your inbox.

  • Never drop a promise.

    You said you'd send it by Friday. It heard you, and it reminds you before it's late.

  • Switch AIs without starting over.

    Use ChatGPT today and Claude tomorrow. Both already know you.

  • Family life, kept private.

    School pickups, appointments, the kids' friends' names. In a private drawer that opens only with your face or fingerprint.

  • Freelance and client work.

    What you agreed with each client, and the email that proves it.

  • Looking for a job.

    Every recruiter, every interview, every follow-up. And exactly what you said about salary.

  • Remember what people told you.

    Birthdays, allergies, the name of their new dog. The small things that matter.

Promise · you oweDue Thu

Send Sam the budget

Sam LeeBudget 2027Work
“I’ll send you the budget by Thursday.”Your email to Sam · Mon 4:12pm

SecondMemory

Claude wants to read Private · Health

To answer: “When is my next check-up?”

Just this once · 1 memory

Approve with Face ID or fingerprint

Don’t allow

A promise, with the email it came from. And the question your phone asks before any AI sees something private.

For people who code

Your agents forget. Your memory doesn’t.

Coding agents start every session from zero. So they re-read your files, re-ask your conventions, and suggest the thing you ruled out last week. SecondMemory gives every agent the same memory.

  • Stop re-explaining your project.

    Your stack, your conventions, how to run the tests. Every agent starts out knowing.

  • Decisions stay decided.

    Why you picked one approach over another, who ruled what out, and when. No more agents re-proposing the thing you rejected last week.

  • Shorter sessions. Same knowledge.

    Start fresh as often as you like. What matters carries over in memory, not in a giant, expensive context.

  • One memory across every tool.

    Claude Code in the morning, Cursor after lunch, a chat assistant in between. They all see the same memory.

  • Your taste, everywhere.

    How you name things, how you like reviews written, how your commit messages read.

  • Work stays work.

    Client code and your own side projects live in separate drawers, with separate permissions.

Your coding agent

Why did we drop Redux here?

Checked SecondMemory · 3 memories

You ruled it out on 12 Aug: the store was only used for two forms. Use React context and keep state local.

Source: email to Sam, Mon 4:12pmWork

See what you could save.

Move the sliders to match a month of your usage, or start from one of ours.

Start from
10M

Fresh tokens you send that aren't cached

3M

What the model writes back

400M

Context re-read on every turn

30M

Context stored in the cache

Your AI bill today

$330/ month

You could save up to*

$129

a month, or $1,542 a year.

39%off your AI bill*

Input tokens
−$12.00
Output tokens
−$9.00
Cache reads
−$40.00
Cache writes
−$67.50

* An upper-range estimate, not a quote. It assumes SecondMemory lets you work in shorter sessions that carry about half as much context on every turn (cache reads −50%, cache writes −45% after the cost of starting fresh sessions), that you paste in and re-read 30% less (input −30%), and that your agents spend 15% less output looking around for things they already decided. Prices are Anthropic’s public list prices for Claude Opus 5.5 on 27 September 2026. Your savings depend on how you work; the app shows the savings it measures for you.

For teams and companies.

One memory for the whole team. Everyone’s private life stays private.

  • A memory the whole project shares.

    Decisions, owners and deadlines, each with the email or meeting it came from.

  • New starters catch up in a day.

    “Why did we do it this way?” gets a real answer, with the source.

  • Customer promises don't slip.

    What your team promised which customer, and when it's due.

  • People leave. Knowledge stays.

    Work memory stays with the company. Their private memory goes with them.

  • Every AI tool, your rules.

    You choose which AI tools can use company memory, and what they can see.

  • Engineering spends less on AI.

    Shorter sessions and less re-reading, across every engineer on the team.

SecondMemory Enterprise · AdminAudit log on
MemoryWhoLocked byAdmin view
Project Atlas42 peopleYour company’s keyShared
Sales playbook18 peopleYour company’s keyShared
Leadership6 peopleYour company’s keyRestricted
Employees’ private memoriesEveryoneEach person’s phoneNot visible

Your company holds the key to its shared memories. Nobody at the company can open an employee’s private ones.

SecondMemory Enterprise

Straight answers.

Is this another AI?
No. Keep using ChatGPT, Claude, Gemini, whichever you like. SecondMemory is the memory underneath them. Think of it as the part that is you, sitting outside all of them.
Does it replace my to-do list?
For promises, mostly. It catches the ones you never got around to writing down, because it heard them in the email or the meeting.
Do you read my email?
Only the sources you connect, and only to pull out facts and promises. We do not train AI on it. We do not sell it. You can disconnect any source and it stops.
What if I want to leave?
Download the lot and go. Free, any time. It is an ordinary file, so other tools can read it.
Who is it for?
Anyone who uses more than one AI, or who keeps promising people things in email and meetings. If you have ever typed your job description into a chat window twice, it is for you.
How do I get in?
SecondMemory is invite only. Request an invite and we’ll email you when your spot is ready.

Rent the AI. Own the memory.

Request an invite