# Korely Memory

> Korely Memory is a freemium MCP server for knowledge and memory. Agent memory that keeps current facts and past values with as-of queries. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Freemium
- **Platforms:** API, CLI, Web, self-hosted
- **Website:** https://korely.ai
- **Repository:** https://github.com/verdana86/korely-memory
- **MCP Registry name:** ai.korely/memory
- **Categories:** knowledge and memory
- **Listing updated:** 2026-11-15

## About

Instead of storing text snippets, Korely extracts typed facts with a validity window. When a new statement contradicts an old one, the earlier fact is closed with an invalid_at date and a pointer to its replacement, not deleted. Asking for context returns the current facts first, so an agent answers with what is true now.

Time handling goes further than most memory stores. An as_of query returns what was true on a past date, and known_as_of returns what the store knew on that date, which differs when a change is reported late. Writes settle asynchronously: raw search works immediately, while typed facts appear after a few seconds, and an events call shows what is still processing.

It is reachable through a hosted MCP endpoint, a REST API, Python and Node SDKs and a CLI, all using one API key. Data is stored in Helsinki; the extraction model step runs on Scaleway in Europe or on Google Gemini in the global region. The vendor publishes benchmark scores on LoCoMo and LongMemEval and says they come from a single run.

## Features

- **Typed bi-temporal facts:** Facts carry validity windows so superseded values are kept, not overwritten.
- **As-of time travel:** Query what was true, or what was known, on a past date.
- **Prompt-ready context:** Returns current facts first with outdated ones left out.
- **SDKs, CLI and REST:** Python and Node SDKs, a CLI and REST all share one key.
- **Audit log:** Records who wrote each memory and what an agent read, without content.
- **Per-user erase:** Delete everything about one end user with one call.

## Good for

- Track a customer's plan changes over time
- Answer what a user's address was on a past date
- Give a support agent only current facts
- Import dated history and query it by period

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http memory https://api.korely.ai/agent/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://api.korely.ai/agent/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "memory": {
      "url": "https://api.korely.ai/agent/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "memory": {
      "type": "http",
      "url": "https://api.korely.ai/agent/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "memory": {
      "serverUrl": "https://api.korely.ai/agent/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "memory": {
      "type": "http",
      "url": "https://api.korely.ai/agent/mcp"
    }
  }
}
```

## FAQ

### Where is data stored?

In Korely's cloud in Helsinki, Finland; the model step runs in the region you pick per project.

### Can I self-host it?

Yes. The code is Apache 2.0 on GitHub, with setup and support under a separate agreement.

### What counts as a write?

One write per 6,000 characters of stored text, however many facts are extracted.

### What happens over the limit?

There is a 10% grace, then requests return 429 until reset, top-up or upgrade; there is no overage billing.

### Is Korely Memory free?

Korely Memory has a free plan, with paid plans for more usage or features.

### How do I add Korely Memory to Claude Code?

Run this command in your terminal: claude mcp add --transport http memory https://api.korely.ai/agent/mcp

### What are the best alternatives to Korely Memory?

The closest alternatives to Korely Memory are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.

## Alternatives

- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Nautilus Compass Memory](https://ai-powered.com/mcp/nautilus-compass-memory): Nautilus Compass Memory is an open-source MCP server for knowledge and memory.
- [note1](https://ai-powered.com/mcp/note1): note1 is an MCP server for productivity.
- [MedXpert 图书馆连接器（通用版）](https://ai-powered.com/mcp/medxpert): MedXpert 图书馆连接器（通用版） is an open-source MCP server for knowledge and memory.
- [Context Nest](https://ai-powered.com/mcp/context-nest): Context Nest is a freemium MCP server for knowledge and memory.
- [Tulsa Ecosystem Map](https://ai-powered.com/mcp/tulsa-ecosystem-map): Tulsa Ecosystem Map is a free MCP server for knowledge and memory.

---
Source: https://ai-powered.com/mcp/korely-memory
