MCP-Nextcloud

MCP-Nextcloud is a standalone Model Context Protocol (MCP) server that connects AI assistants to a self-hosted Nextcloud instance.

  • AI
  • Productivity
MCP-Nextcloud is a standalone Model Context Protocol (MCP) server that connects AI assistants to a self-hosted Nextcloud instance. It exposes Nextcloud data through a secure API so Large Language Models (LLM) like Claude, GPT, and Gemini can create notes, manage calendars, organize contacts, and work with files via natural-language prompts from external clients such as Claude Code or IDEs. It runs independently of Nextcloud in Docker, a VM, Kubernetes, or locally, and is aimed at homelabbers who want fine-grained access to Nextcloud apps from their own AI tools instead of using Nextcloud's built-in Assistant.

Platform availability

Available on 1 of 10 platforms

  • Unraid
  • TrueNAS (not listed)
  • Umbrel (not listed)
  • ZimaOS (not listed)
  • Proxmox (not listed)
  • Helm (not listed)
  • Mac (not listed)
  • Windows (not listed)
  • Linux (not listed)
  • Any server (not listed)

Health score?A 0-100 score computed nightly from four factors: maintained (40%: recent commits and steady activity), popular (25%: GitHub stars, log scale), easy to install (20%: platforms, install notes, setup guides, Docker image) and light to run (15%: minimum RAM, ARM support). When the source code has been checked, a fifth factor, security check (20%: automated scan for committed secrets, unsafe desktop-app settings, known-vulnerable dependencies and project hygiene; any serious finding caps it at 20), is added. Factors without data are left out and the rest are rescaled.

58/100 Fair

Maintained
100last commit 6 days ago
Popular
52413 GitHub stars
Easy to install
351 install platform, install notes, Docker image
Light to run
502 GB minimum RAM
Security check
103 serious findings, 25 minor, CI, tests

Availability: platform listings confirmed today, links checked 20 days ago

GitHub stars
413
Open issues
100
Last commit
2026-11-15
Latest release
v0.211.1
License
AGPL-3.0
Activity
steady

Checked 6 days ago - source: GitHub (cbcoutinho/nextcloud-mcp-server)

Resources & Compatibility

Minimum requirements: 2 GB RAM, 1 CPU core(s), 40 GB disk.

Checked 54 days ago - source: Docker Hub image tags and published docs

First-install notes

  • Install the MCP-Use CLI first Run `npm install mcp-use` or install the CLI globally with `npm install -g @mcp-use/cli` before trying start or build commands. Without this, your terminal won’t recognize `mcp-use` and the install will stop right away. source
  • Pick the right port and host at start Use `mcp-use start` for local use, or add flags like `--port 8080` and `--host 0.0.0.0` when you need a different port or access from another machine. If the host is too open or the port is wrong, your client may not be able to connect or you may expose the app more than intended. source
  • Build before running with Inspector Run `mcp-use build` and then `mcp-use start --with-inspector` to get a built server plus a built-in debugging view. This is a useful first-run check because it helps you see whether the service is actually working before you point tools or automation at it. source

Fetched 53 days ago - each note links to its source

Alternatives

No alternatives collected yet.

Common questions

Can I use it with Claude, GPT, or Gemini, and what Nextcloud actions are available?

It lets Large Language Models like Claude, GPT, and Gemini interact with your Nextcloud data through a secure API. It can create notes, manage calendars, organize contacts, and work with files.

What Login Flow v2 issues should I look for in multi-user setups?

Multi-user setups can run into Login Flow v2 issues involving app-password storage, provisioning loops, and OAuth issuer problems.

What should I check before upgrading from v0.57.x?

Use the Configuration Migration Guide for help when upgrading from v0.57.x.

Does self-hosting the Nextcloud instance for this server increase security exposure?

One community thread says self-hosting can increase the chance of someone hacking the machine and grabbing your data. It contrasts that with large cloud providers, which have staff monitoring malicious behavior, while self-hosters are left reviewing logs and trusting their own setup.

Answers sourced from github.com, github.com, www.reddit.com

Community

Collected 43 days ago - public community threads