Nextcloud Task MCP Server
Enables managing Nextcloud tasks through natural language, including creating, updating, completing, and deleting tasks.
README
nextcloud-task-mcp
An MCP server that manages tasks (VTODOs) in a self-hosted Nextcloud instance over CalDAV. Connect it to Claude as a custom connector to create, list, update and complete Nextcloud tasks using natural language.
Built with FastMCP on the Streamable HTTP transport, and the
caldav library for talking to Nextcloud.
Documentation:
- Deployment guide — Ubuntu LXC + Tailscale + systemd + Claude connector setup
- Tool reference — all tools with parameters, examples and error messages
- Architecture — module layout, request flow, design decisions
How it works
- One CalDAV connection is opened at startup and reused for every request (no reconnect-per-call).
- Every incoming MCP request must carry a static bearer token
(
Authorization: Bearer <token>), checked by a middleware before any tool or CalDAV logic runs. - The server binds to a local HTTP port only (e.g.
127.0.0.1:8000). It does not handle TLS itself - in the intended deployment,tailscale serveterminates TLS in front of it and only devices on your tailnet can reach it. - CalDAV/network failures (auth errors, timeouts, missing task lists/UIDs, ...) are caught and turned into short, clean error messages - no raw stack traces are ever returned to the MCP client.
Setup
Requires Python 3.10+ and uv.
uv sync
cp .env.example .env
# edit .env with your Nextcloud CalDAV URL, an app password, and a bearer token
Generate a Nextcloud app password under Settings → Security → Devices & sessions (never use your account password). The CalDAV URL is your DAV endpoint root, typically:
https://<your-nextcloud-domain>/remote.php/dav/
Generate a random bearer token, e.g.:
python3 -c "import secrets; print(secrets.token_urlsafe(32))"
Run the server:
set -a; source .env; set +a
uv run nextcloud-task-mcp
It listens on MCP_HOST:MCP_PORT (default 127.0.0.1:8000) at the /mcp path, using the
Streamable HTTP transport. Point your Claude custom connector at
http://<host>:<port>/mcp with the configured bearer token.
Tools
All tool parameter names match the field names below exactly (including German field
names with umlauts, e.g. priorität, fällig_datum) - this is the literal MCP tool
schema Claude calls.
list_task_lists()
Returns all available Nextcloud task lists as {"name": ..., "url": ...} dicts
(display name and internal CalDAV URL/ID).
list_tasks(list_name, nur_offene=True)
Returns tasks in a list. nur_offene=True (default) excludes completed tasks. Each task
is a dict with: uid, titel, start_datum, fällig_datum, priorität,
fortschritt_prozent, status ("offen" / "erledigt"), ort, url, tags,
notizen, übergeordnete_uid (parent task UID, or null if not a subtask).
create_task(liste, titel, ...)
Creates a task. Required: liste, titel. Optional fields and their CalDAV mapping:
| Parameter | CalDAV property | Notes |
|---|---|---|
start_datum |
DTSTART |
ISO 8601 date or datetime |
fällig_datum |
DUE |
ISO 8601 date or datetime |
priorität |
PRIORITY |
"hoch"→1, "mittel"→5, "niedrig"→9 |
fortschritt_prozent |
PERCENT-COMPLETE |
0-100 |
ort |
LOCATION |
|
url |
URL |
|
tags |
CATEGORIES |
list of strings |
erinnerungen |
VALARM |
see below |
notizen |
DESCRIPTION |
|
sichtbarkeit |
CLASS |
"öffentlich"→PUBLIC, "privat"→PRIVATE, "vertraulich"→CONFIDENTIAL |
übergeordnete_aufgabe |
RELATED-TO;RELTYPE=PARENT |
UID of an existing task; makes this task its subtask |
Reminders (erinnerungen): each entry is either a relative RFC 5545 duration (e.g.
"-P1D", "-PT1H") or an absolute ISO 8601 datetime. Relative reminders trigger before
fällig_datum if set, otherwise before start_datum; a relative reminder without either
date raises an error. Absolute reminders without a UTC offset are interpreted as UTC (per
RFC 5545, VALARM triggers must be in UTC).
update_task(list_name, task_uid, ...)
Same fields as create_task, all optional except task_uid. Only fields you pass are
changed; everything else on the task is left untouched. Passing erinnerungen replaces
all existing reminders on the task.
complete_task(list_name, task_uid)
Sets STATUS:COMPLETED, PERCENT-COMPLETE:100, and a COMPLETED timestamp.
delete_task(list_name, task_uid)
Permanently deletes the task.
Testing
Unit tests mock the caldav library entirely - no network access, no real Nextcloud
instance required:
uv sync # installs the dev group (pytest, ruff) by default
uv run pytest -q
Integration tests exercise the full flow against your real Nextcloud instance (create, list, update, complete, delete a task in a disposable test list). They're skipped by default. To run them:
export RUN_INTEGRATION_TESTS=1
export NEXTCLOUD_CALDAV_URL=... NEXTCLOUD_USERNAME=... NEXTCLOUD_APP_PASSWORD=...
export INTEGRATION_TEST_LIST="Test" # an existing task list; tasks are created/deleted in it
uv run pytest -q
License
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。