A production-ready MCP server for Nextcloud, providing 90+ tools for AI assistants to directly manage files, notes, calendars, and boards with Docker/Kubernetes deployment support.
Nextcloud MCP Server is a high-performance Python middleware that empowers AI assistants (Claude, GPT, Gemini) to deeply integrate with Nextcloud via the Model Context Protocol (MCP).
Core Value#
Solves the challenge of securely connecting AI to private cloud data. Users can grant AI read/write access to specific application data without exposing full admin credentials.
Feature Highlights#
Application Support (8 Apps, 90+ Tools)#
| App | Tools | Core Capabilities |
|---|---|---|
| Notes | 7 | Full CRUD, keyword search, semantic search |
| Calendar | 20+ | Events/todos, recurring events, attendee management |
| Files (WebDAV) | 12 | File system access, OCR/document processing |
| Deck | 15 | Boards, stacks, cards, tags, assignments |
| Contacts | 8 | CardDAV support, address book management |
| Cookbook | 13 | Recipe management, URL import (schema.org) |
| Tables | 5 | Nextcloud Tables row operations |
| Sharing | 10+ | Create and manage shares |
Advanced Features#
- Semantic Search (Experimental): Vector-based similarity search for Notes, Files, News items, Deck cards
- Document Processing: OCR and text extraction for PDF/DOCX/images (PyMuPDF + Pillow)
- Multiple Transports: SSE, HTTP, streamable-http
- Observability: Prometheus metrics, OpenTelemetry integration
Typical Use Cases#
- Note Management: "Create a note named 'Meeting Notes' with today's agenda"
- Recipe Import: "Import recipe from https://example.com/recipe"
- Schedule Management: "Schedule a team meeting next Tuesday at 2 PM"
- File Operations: "Create folder 'Project X' and move all PDFs there"
- Semantic Search: "Find notes related to machine learning"
Deployment#
Docker (Recommended)#
docker run -p 127.0.0.1:8000:8000 --env-file .env \
ghcr.io/cbcoutinho/nextcloud-mcp-server:latest
Local Installation#
uv pip install nextcloud-mcp-server
Authentication#
- Recommended: Basic Auth + App Password (production-ready)
- Experimental: OAuth2/OIDC (requires user_oidc patch)
Key Tech Stack#
- Python >= 3.11
- MCP Python SDK, Starlette (ASGI), httpx
- Pydantic v2, SQLite + Alembic
- Qdrant, FastEmbed, BM25 (semantic search)
Project Maturity#
- 1900+ commits, 323 tags, current version 0.64.3
- Full test suite + CI/CD (GitHub Actions)
- 22+ Architecture Decision Records (ADR)
- License: AGPL-3.0-only