Skip to content
MCP server

Korpus

Shared Markdown memory that teams and their AI agents read and write, with an audit trail

Install
Pricing
Freemium
Platforms
Web, API, self-hosted
Runs
Remotely
Version
6.0.0
Listing updated
2 days ago

About Korpus

Korpus is a freemium MCP server for knowledge and memory. It runs as a hosted remote server, so there is nothing to install locally.

Korpus stores memory as bundles: folders of Markdown concepts for one subject, private or shared. A concept is one document at a path with a type, tags, links and a revision number that counts every edit. Links are typed, such as supersedes or depends on, and each bundle names its own kinds. README and AGENTS concepts tell people where to start and tell agents how to behave in that bundle.

People and agents work side by side. Agents such as Claude, ChatGPT, Cursor or a CI job connect over MCP, each acting for one person and limited to the bundles that person grants, with read, write or path-level access. Every change is stamped with who wrote it, and each write names the revision it read so nobody overwrites blind.

It is hosted in Frankfurt (eu-central-1) and runs serverless on AWS. On the Business plan the same stack can be deployed into your own AWS account, with Microsoft Entra sign-in. Free is for private non-commercial use, and Pro is 10 euros a month or 100 a year.

Install

Run in a terminal

claude mcp add --transport http korpus https://mcp.fra.korpus.cloud/

Add to Settings → Connectors → Add custom connector

https://mcp.fra.korpus.cloud/
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "korpus": {
      "url": "https://mcp.fra.korpus.cloud/"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "korpus": {
      "type": "http",
      "url": "https://mcp.fra.korpus.cloud/"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "korpus": {
      "serverUrl": "https://mcp.fra.korpus.cloud/"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "korpus": {
      "type": "http",
      "url": "https://mcp.fra.korpus.cloud/"
    }
  }
}

What it does

Bundles of concepts
Folders of Markdown with types, tags and typed links.
Scoped agent access
Agents get read, write or path-limited access per bundle.
Revision-checked writes
Each write names the revision it read, preventing blind overwrites.
Attribution on every change
Edits show whether a person or their agent wrote them.
Catalog templates
Start from ready bundles such as incident response or security audit.
Own-cloud deployment
Business plan can run in your own AWS account and region.

Good for

  • Give a team's Claude Code the same decisions and runbooks
  • Keep a per-client memory and hand it over at project end
  • Let a CI job write findings that people can review
  • Maintain architecture decision records that agents follow

Pricing

Freemium

Free plan; Pro 10 EUR/month or 100 EUR/year; Business on request

Prices change. Check korpus.cloud before you buy.

Korpus FAQ

What does it cost?

Free for private, non-commercial use; Pro is 10 euros a month or 100 a year; Business is on request.

Where is data hosted?

In Frankfurt (eu-central-1), or in your own AWS account on the Business plan.

Do agents count as users?

No. The pricing page says agents never count as users.

Which agents can connect?

Claude, ChatGPT, Cursor or a CI job over MCP.

Is Korpus free?

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

How do I add Korpus to Claude Code?

Run this command in your terminal: claude mcp add --transport http korpus https://mcp.fra.korpus.cloud/

What are the best alternatives to Korpus?

The closest alternatives to Korpus are Codebase Memory, XERJ, Ha MCP, Claude Code Ultimate Guide and Nautilus Compass Memory.

Alternatives to Korpus

  1. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  2. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  3. Ha MCP

    MCP server

    MCP server for managing and automating Home Assistant from AI assistants

    Runs locally
    • 8 tools
    • Open source
    • 7.9K GitHub stars
    • Productivity
  4. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  5. Local-first long-term memory layer for AI agents, served over MCP

    Not checked yet
    • 5 tools
    • Remote
    • Open source
    • 1.3K GitHub stars
    • Knowledge and memory
  6. Concord MCP

    MCP server

    Local-first MCP server for messaging and shared work-state between coding agents

    Runs locally
    • Freemium
    • 1.3K GitHub stars
    • Developer tools