What MCP is

MCP (Model Context Protocol) is the open standard Claude uses to talk to external services. An MCP server provides Claude with tools such as "search my notes" or "read this document", and Claude calls them on its own when a question fits. More background is in the MCP fundamentals article; this paragraph is enough for setup.

Prerequisites

• A Memexa account (create one free, the Free plan is enough)
• Claude Desktop or Claude Code

You don't need API keys or config files. Sign-in happens through OAuth in the browser, and the address is the same for every client:

https://brain.memexa.ai/mcp

Connect Claude Desktop

1. Open Settings in Claude Desktop (gear icon, bottom left).
2. Choose the Connectors tab.
3. Click Add custom connector.
4. Paste https://brain.memexa.ai/mcp as the URL and confirm.
5. Sign in with your Memexa account in the browser window that opens. The sign-in is permanent.

Afterwards, "memexa" appears in your connector list.

Connect Claude Code

One command in the terminal:

claude mcp add --transport http --scope user memexa https://brain.memexa.ai/mcp

--scope user makes the connection available in all projects, not just the current folder. On first access the browser opens for sign-in. After that the connection runs in the background.

Check that it works

Ask Claude a question only your memory can answer:

"What do you know about me?"

If Claude calls tools like search_brain or list_recent and answers with content from your Memexa, the connection is up. In Claude Code, /mcp shows the status of every connected server.

What Claude can do afterwards

Claude now remembers things across sessions: say "Remember: release moved to the 14th" and the note is still there next week, in ChatGPT or Cursor too if you connect those as well. Uploaded PDFs and scans go through text recognition automatically; ask Claude about one and the answer names the source, such as "contract, page 4". New notes also link themselves to existing ones, so Claude can navigate the graph from one piece of information to the next.

Common problems

The connector shows up, but Claude doesn't use it

Claude only calls tools when the question sounds like it needs them. Be explicit at first: "Check my Memexa for …". After the first few hits, Claude starts using the memory unprompted.

Sign-in fails or hangs

Usually a browser issue: disable the pop-up blocker, then remove and re-add the connector. There is nothing to enter in Claude itself; sign-in happens entirely in the browser.

"Server disconnected" in Claude Code

Run claude mcp list. If memexa shows "failed", claude mcp remove memexa followed by the claude mcp add from above almost always fixes it. If the error persists, check whether a corporate firewall filters outgoing HTTPS connections.

Claude can't find a document

Freshly uploaded documents need a moment for text recognition and linking. You can see the status in the dashboard. Once the document is marked as processed, Claude finds it.

The protocol behind them — MCP, the open standard AI clients use to talk to external services — has its own article. This one stays on the practical side: the server.

What an MCP server does

An MCP server is a service that offers tools to an AI. Each tool is a named capability with a clear job: search my notes, read this file, create a GitHub issue. The client — Claude, ChatGPT, Cursor — receives the list of tools when it connects, and decides on its own when to call which one.

That's the whole trick. The server doesn't need to understand AI; it just answers requests. The model doesn't need to understand the service; it only sees described tools. Which is why the same server works with every client that speaks the protocol.

Which servers exist

The directories are full, but almost everything falls into four groups:

• Service connectors. GitHub, Notion, Slack, Google Drive, databases. The server translates between the AI and the service's API. Useful when you already work there.
• System tools. File system, terminal, browser automation (Playwright, for instance). Mostly interesting for developers building AI agents or letting AI write code.
• Search and retrieval. Web search, documentation lookup, scraping. They give the model current information that wasn't in its training data.
• Memory servers. They store what you tell your AI and make it findable in later sessions. The most interesting group if repeating your context forever annoys you — and the one with the biggest quality differences.

For everyday use you don't need ten servers. Most people are fine with one or two: one for work (GitHub, Notion) and a memory.

How to connect a server

Setup follows the same pattern in every client — enter the address, sign in once:

• Claude Desktop and ChatGPT: Settings → "Connectors", paste the server's address. Sign-in opens in the browser.
• Claude Code: one terminal command, e.g. claude mcp add --transport http memexa https://brain.memexa.ai/mcp.
• Cursor: one line in the MCP config file.

After that, the server shows up in the client's tool list and the AI uses it on its own whenever a question calls for it. What this looks like for Memexa is on the setup page — with copy-paste configuration for every client.

Local or hosted?

Many servers in the directories are small programs you start yourself — via npx in a terminal, configured in a JSON file. That's fine for developers and the most common hurdle for everyone else: the server only runs while your machine runs, you handle updates yourself, and on your phone it doesn't exist.

Hosted servers invert this. They run as a service on the web, you connect with a URL and a browser sign-in (OAuth), and the same state is reachable from every device and every client. In exchange, you hand your data to someone else — which brings us to the real question.

What to look for when choosing

The protocol is open; the quality behind it is not regulated. Three questions separate the solid servers from the risky ones:

1. Where does the data live? With local servers, on your machine. With hosted ones: which country, under which law? A server that doesn't answer this has already answered it.
2. What is the server allowed to do? A tool that can delete files or send mail deserves more suspicion than one that only reads. Don't connect more permissions than the task needs.
3. Can you leave? If the service disappears or you want to switch: is there a complete export in an open format?

Memexa's answers: EU hosting, only you, one-click Markdown export.

Frequently asked questions

Do I need to know how to code?

For hosted servers with OAuth sign-in: no — entering an address and signing in is all. Local servers from the community directories usually assume a terminal doesn't scare you.

How many servers can I connect at once?

Technically many; practically, restraint pays off. Every server makes the tool list longer, and the model walks it on every request — too many similar tools make its choices worse, not better.

What separates memory servers from each other?

Most are scratchpads: the AI stores text snippets it deems worth keeping and searches them later. That already helps. A hosted memory like Memexa goes further — you also throw in PDFs, scans and mail, everything is read, linked, and returned with its source, down to the page number. The difference shows three months later, when you want to know where a fact came from.

That's the problem behind an unassuming acronym you now see everywhere: MCP.

The short answer

MCP stands for Model Context Protocol — an open standard that defines how AI applications talk to external data sources and tools. Anthropic published it in late 2024 and deliberately built it as an open protocol rather than an exclusive feature. By now virtually every relevant AI client supports it: Claude, ChatGPT, Cursor, and many more.

The common analogy holds up: MCP is the USB-C port for AI. Before USB-C, every device needed its own cable. Before MCP, every connection between an AI and a data source needed its own hand-built integration. With MCP there is one plug that fits everywhere: a service offers its capabilities once as an "MCP server" — and every AI client that speaks the protocol can use them.

How it works in practice

Behind the protocol sits a simple division of roles:

• The client is the AI application you talk to — Claude, ChatGPT, Cursor.
• The server is a service that can do or knows something — it offers the AI "tools": search my data, save this note, read this document.
• You connect the two once, usually with an address and a short browser sign-in.

From then on the important part happens automatically: when you ask your AI a question that needs external knowledge, it calls the right tool, gets the result back, and works it into its answer. All you notice is that the answer is suddenly correct.

What does this have to do with memory?

Out of the box, every AI session is an island. The built-in memory features of ChatGPT and Claude help to a point — but they belong to the respective tool: what ChatGPT has learned about you stays in ChatGPT. Switch models and the memory stays behind.

MCP inverts that relationship. If your memory is its own MCP server, it no longer belongs to any client — it belongs to you. Every AI you connect reads and writes the same body of knowledge:

You tell Claude on Tuesday: "Remember: release moved to the 14th." You ask ChatGPT on Friday: "When was the release again?" — and get the answer. With the source.

That's the difference between a memory feature and memory as its own layer: models come and go, the knowledge stays in one place — searchable, linked, and exportable.

Do I need technical skills for this?

Fewer than the word "protocol" suggests. For most clients, setup takes minutes and always follows the same pattern: enter the MCP server's address, sign in once in the browser, done. In Claude Desktop and ChatGPT the menu is called "Connectors", in Cursor it's one line in a config file, in Claude Code a single command.

For Memexa it looks like this — including copy-paste configuration for every client.

Frequently asked questions

Does MCP cost anything?

The protocol itself, nothing — it's an open standard. Costs can appear on two sides: some clients only allow custom connectors on paid plans (ChatGPT, for instance), and the MCP server you connect may be a paid service.

Is it secure?

MCP itself only defines the language between client and server — how safe your data is depends on the service behind it: where it hosts, how it secures access, what it does with your content. The questions to ask any MCP service: Where does the data live? Who can read it? Can I leave any time? (Our answers: EU, only you, yes — as Markdown.)

Which clients support MCP?

As of today: Claude (Desktop, Code, Web), ChatGPT, Cursor, and a fast-growing field of other tools. Since the standard is open, practically every new AI application ships with MCP support — which means an MCP-based memory also works with the clients that don't exist yet.

There are many MCP servers — what sets them apart?

Hundreds. Most memory servers are developer tools: locally installed, self-maintained, storing whatever the AI deems worth keeping. A hosted memory like Memexa sits one layer up: you also throw in documents, PDFs and emails, everything is read via text recognition, linked — and every answer cites its source, down to the page number.