# Shared memory for your team’s Cursor

MemScope is a shared memory that your team’s Cursor agents read and write over a remote MCP server. What one person’s Cursor learns in a chat can be recalled by a teammate’s Cursor, if that teammate may see it.

## The problem

A Cursor agent settles a question in chat: which endpoint to call, which library is out. When the chat ends, the answer goes with it, and the next person’s agent asks the same question.

Rules fix part of that. They are written down once and apply to everyone they target. Writing a rule is a deliberate step, though, and a rule reaches everyone it applies to, so a decision meant for one team has no natural home.

## How MemScope fits

Cursor connects to MemScope as an MCP server. Agents save a decision or convention as they work, and another person’s Cursor searches those memories before it starts, getting back the ones that person may read. AI suggests labels such as sensitivity and topic, but a deterministic policy decides access. Organization admins get no read bypass, MemScope staff see XXX unless the customer grants time-limited, audited support access, and every change is audited.

## Install in Cursor

Add the server to Cursor's MCP config. Where: ~/.cursor/mcp.json (all projects) or .cursor/mcp.json (this project).

```json
{
  "mcpServers": {
    "memscope": {
      "url": "https://mcp.memscope.ai/mcp"
    }
  }
}
```

Open Cursor Settings → MCP and click Connect / Needs login next to memscope.

New to MemScope? Create an account at https://memscope.ai/signup, choose your organization and approve. All clients: https://memscope.ai/agents.md

## What Cursor rules do

Cursor has Project Rules (files in .cursor/rules, version-controlled with the repository), User Rules (personal preferences across projects) and Team Rules, which Team and Enterprise plan administrators manage from the Cursor dashboard and can mark as required. Cursor also supports AGENTS.md. These are standing instructions that a person or an administrator writes. Those docs describe no per-person or per-group targeting (rules are scoped by file pattern) and do not describe a memory feature.

Source: Cursor docs: Rules, https://cursor.com/docs/context/rules (checked November 2026).

Compare: [MemScope vs Cursor Team Rules](https://memscope.ai/compare/cursor-team-rules)

## Example: same question, two answers

An illustration. Two memories exist: one shared with the whole organization, one limited to the Platform team.

- Shared with everyone: New callers use the v2 auth endpoint.
- Platform team only: The v1 auth endpoint is being retired; questions about it go to the Platform on-call.

Two people ask Cursor’s agent: “Which auth endpoint should the new settings page call?”

- A teammate on Platform: Gets the v2 guidance and the note about v1.
- A teammate on another team: Gets the v2 guidance. The Platform-only note is not returned to them.

## Who it is for

Teams whose developers work in Cursor, alone or alongside Claude Code or Codex, and who want one shared memory across those tools instead of a rule file per tool.

## Price

$15 per connected user per month after a 60-day trial with 3 connected users. You pay only for people who connect an AI client.

- [Start free](https://memscope.ai/signup)
- [All integrations](https://memscope.ai/integrations)

---

Web page: https://memscope.ai/cursor · All facts for AI agents: https://memscope.ai/llms.txt
