Context Engine

npx add-mcp 'https://app.contextengine.com/mcp'

Invoicing app

Arrived today4 memories

Every AI tool you use starts from nothing.

Claude Code forgets what you decided yesterday. Cursor does not know what ChatGPT learned this morning. You explain the same project and the same rules again in every tool, every session. And what one person’s AI learns stays in that person’s chat history, where the rest of the team can’t see it.

Context Engine is the memory your AI tools and your team share. It is a hosted service at app.contextengine.com that you sign in to with your email. You connect your AI tools once, and as they work they save what matters into it: decisions, notes, people, customers and files. The next session starts with all of that, and anyone on your team can read it.

• Your AI tools save to it and read from it through MCP, the standard they already speak.
• You and your team see every memory, who saved it and when, ask it questions in plain words and add files yourselves.
• It is yours: export all of it as Markdown any time, or delete it.

Connect your AI tools once, and they remember from then on.

1

Connect your AI tools

Pick the app you use. ChatGPT and Claude take a few clicks, and Claude Code takes one command. Each connection reaches only the engines you choose, and there is no key to copy or keep safe.

Saves to Invoicing app

$ claude mcp add --scope user --transport http ce-invoicing \
    https://app.contextengine.com/mcp/c/k7f2q9

2

They remember as they work

As you work, your AI tools save what matters: a decision, a customer’s request, a new contact. Each memory has a type, shows which tool saved it and keeps its history. You can also drop in files, a whole folder or a.zip.

MemoriesInvoicing app

• decisionRetry card payments at most three times
• customerContoso wants invoices grouped by cost centre
• personMarta Reyes, Contoso’s finance contact
• noteStripe sends a second webhook when a retry succeeds
• fileQ3 pricing review.pdf

3

The next session picks up where the last one stopped

When a tool starts, it gets a short brief: what each engine is for, what happened recently and what changed since its last visit. If it needs more, it searches the memory, matching meaning as well as exact words. You read the same memory in the app, so you can always see what your tools know.

Claude Code · new session

Brief

Invoicing app: code, customers and decisions for our invoicing product.

Since you were last here

• Cursor decided card payments retry at most three times.
• ChatGPT noted Contoso wants invoices grouped by cost centre.
• Marta Reyes is Contoso’s new finance contact.

Picking up the retry change. I’ll cap retryPayment() at three and group Contoso’s invoices by cost centre.

An engine is a memory with a purpose.

Say in a sentence or two what an engine is for. Your AI tools read that purpose to decide what to save where, so the client’s decisions land in the client’s engine and your own notes stay yours.

• As many as you need. One for yourself, one per client or project, one for a team.

• Walls between them. A connection reaches only the engines you tick, each at Can read or Can read and save.

• Labelled answers. A search across engines names the engine every result came from.

• Move without losing history. A memory moved to another engine takes its history with it.

• S Selin Consulting

• A Acme

• G Globex

Claude

AcmeCan read and save

Selin ConsultingCan read

GlobexNot reached

It stays organised without you sorting it.

Context Engine writes summaries, links related memories and catches duplicates, so what was saved last month is still useful this month, to your team and to your tools.

Ask your memory

Ask in plain words. Every answer names the memories it came from.

What did we decide about card retries?

Card payments retry at most three times, down from five after August’s disputes. A card flagged as stolen or fraudulent is never retried.

decisionRetry card payments at most three times decisionNever retry a card flagged as stolen

Summaries, written for you

A short account of each day, rolled into weeks and months, from what was saved.

Card retries took the day: after August’s disputes the team cut retries from five to three, and Jon drafted the runbook change. Contoso asked for invoices grouped by cost centre.

Linked

People, customers, decisions and files connect to each other, and the map shows how.

Contoso Marta Reyes Cost centres Retry rule Q3 review.pdf

No duplicates by accident

When a tool tries to save something already there under the same name, it is pointed to that memory to update instead.

Marta Reyes is already saved

History on everything

Every change records who made it, with which tool, and what it was before.

Jon edited “Retry card payments at most three times”

The AI tools you already use.

Sign in from the apps below, or connect anything that speaks MCP or HTTP with a key.

• ChatGPT
• Claude
• Claude Code
• Cursor
• Your own agent

For one person, for every client, for the whole team.

If you work for several clients

One engine per client, and nothing crosses.

Give each client an engine and a connection of its own. In Claude Code, a project-scoped connection keeps each client’s repository to that client’s memory.

Why trust it with your memory.

Your memory belongs to you and is stored apart from every other customer’s. You can see all of it, and take it with you.

It is yours

Export an engine any time as one ZIP of Markdown and JSON that opens without us. Every memory is a plain Markdown file.

Kept apart for real

Every engine gets its own database, its own search index and its own running copy of the service. It never shares a table with anyone else’s data.

Never used for training

Your content is sent to models only to answer you. It is not used to train anything, not sold and not used for advertising.

Staff only with your say-so

Support opens an engine only when you approve a request, for the time it names, and every visit is logged. A legal or safety duty is the one exception, and you are told at once.

You decide what reaches it

See every tool and key that can reach an engine, with its access and when it was last used. Pause it, or cut it off.

Deleting means deleting

Deleting an engine is scheduled a day ahead so you can export first. Then it is gone. Choose the EU at setup to keep its database and files there.

Two-step sign-in is built in. Enterprise puts Context Engine on your own server, for your organisation alone. Read the privacy notice.

Pay per engine, never per person.

People are free on every plan. Each hosted engine has its own plan, set by how many agents connect to it. Enterprise, on your own server, is priced per install.

Free $0forever 1 agent per engine, up to 3 when friends join. 1 GB. No card. Pro $12/month 10 agents and 25 GB per engine. Or $99 a year. Unlimited $39/month Every agent you run and 100 GB per engine. Or $299 a year. Enterprise From$6,000/year On your own server, run by Relote, and single sign-on for your own apps. Updates and support included.

Compare plans How referrals add agents to Free

Asked often.

The rest is in the help, which is public, or write to [email protected].

Is it a chatbot?

No, and it does not run your agents. Your AI tools do the thinking. Context Engine is the memory they share: they save to it as they work and read from it when they start. You can also ask it questions yourself, and every answer names the memories it came from.

Is it for people or for AI tools?

Both. Your AI tools save to it as they work and read from it when a session starts. People use the same memory in the app: browse every memory and who saved it, ask questions and get answers that name their sources, read summaries of each day and week, and drop in files. Your whole team shares it on every plan, at no cost per person.

Which AI tools work with it?

ChatGPT, Claude on the web and the desktop app, Claude Code and Cursor sign in with a few clicks. Anything else that speaks MCP or HTTP connects with a key, so agents and scripts you build yourself can use the same memory.

What is an engine?

An engine is a memory with a purpose: a few sentences, in your words, saying what it is for and what belongs in it. Your AI tools read the purpose to decide what to save there. Keep one for yourself, one per client or project, or one for a team.

Can I keep clients apart?

Yes. Each engine is stored on its own, and each connection reaches only the engines you tick, at Can read or Can read and save. Give each client its own engine and its own connection, and nothing crosses between them.

Where is my data, and who can see it?

Every engine gets its own database, its own search index and its own running copy of the service, on Cloudflare. Your content is sent to models only to answer you and is never used to train anything. Context Engine support opens an engine only when you approve a request, for the time it names, and every visit is logged.

Can I take my memory with me?

Any time. Export gives you one ZIP of Markdown and JSON that opens without us: every memory is a Markdown file, so a text editor or a notes app reads it as it is. Deleting an engine is scheduled a day ahead so you can export first, and then it is gone.

What happens when my memory is full?

Every plan lets your AI tools read and save as much as they need; what an engine’s plan sets is how many agents connect to it and how much memory it holds. You are told before your memory is full. When it is, a new memory is refused with the reason, and nothing you stored is ever deleted.

Who makes Context Engine?

Relote built it and runs it. If you want help putting AI to work across your company, that is Relote’s work, at relote.com. Context Engine itself is the product: sign up, connect your tools, and it works.