kalendia-mcp

kalendia-mcp

An MCP server for Kalendia that enables AI assistants to manage calendar connections, sync rules, scheduling pages, and agenda directly from chat.

Category
访问服务器

README

kalendia-mcp

An MCP server for Kalendia. Drive your calendar sync and scheduling from an MCP client (Claude Code, Claude Desktop): list connections and calendars, read your agenda, and create / delete / run sync rules, all from chat.

Tools (24)

The tool surface mirrors what a user can do in the Kalendia web app (everything reachable by an API token; OAuth-redirect actions like connecting Google/Microsoft or Stripe checkout stay in the web UI, and account deletion is intentionally not exposed).

Reads (read-only):

  • list_connections — connected calendar accounts (Google, Microsoft, iCloud, ICS, Zoom).
  • list_calendars(connection_id) — live provider calendars under one connection (no Kalendia id).
  • list_all_calendars — every persisted calendar across accounts, with the numeric id used by rules/pages.
  • list_sync_rules — all mirror rules.
  • get_agenda(connection_id, calendar_id, from_iso?, to_iso?) — events in a window.
  • list_scheduling_pages / get_scheduling_page(page_id) — booking pages.
  • get_availability(slug, from_iso?, to_iso?) — bookable slots for a page.
  • get_billing / get_billing_overview — plan, usage, limits.
  • get_audit_log(limit?) — recent account activity.

Writes (the client confirms before firing):

  • Sync rules: create_sync_rule, run_sync_rule, delete_sync_rule (destructive).
  • Connections/calendars: connect_icloud, connect_ics, refresh_calendars, discover_calendars, set_active_calendars, rename_calendar, disconnect_connection (destructive).
  • Scheduling pages: create_scheduling_page, update_scheduling_page, delete_scheduling_page (destructive).

Setup

  1. Mint a personal access token in Kalendia: Settings > API tokens. Copy it (shown once; starts with kld_).
  2. Install deps: uv sync.
  3. Configure the env: copy .env.example to .env and set KALENDIA_TOKEN (and KALENDIA_API_URL if testing against a local backend).

Run

uv run kalendia-mcp          # stdio (default)
uv run kalendia-mcp --http   # Streamable HTTP

Register in Claude Code

Add to your MCP config (e.g. ~/.claude.json), filling in the token:

{
  "mcpServers": {
    "kalendia": {
      "command": "uv",
      "args": ["run", "--directory", "/Users/bogdan-georgebrad/Claude/Code/personal/kalendia-mcp", "kalendia-mcp"],
      "env": {
        "KALENDIA_TOKEN": "kld_your_token_here",
        "KALENDIA_API_URL": "https://api.kalendia.io"
      }
    }
  }
}

To test against a local backend instead, set KALENDIA_API_URL to http://localhost:8002 and use a token minted on that backend.

Remote / multi-user (HTTP)

uv run kalendia-mcp --http runs the server over Streamable HTTP as an OAuth 2.0 resource server, so ONE deployed instance serves many users: each request authenticates with the caller's own kld_ token (no per-user env var). It serves OAuth Protected Resource Metadata at /.well-known/oauth-protected-resource (advertising Kalendia as the authorization server), rejects unauthenticated/invalid tokens with 401 + WWW-Authenticate, validates a presented kld_ token against the Kalendia API, and forwards it to act as that user. Config via env: KALENDIA_MCP_HOST / KALENDIA_MCP_PORT (bind), KALENDIA_MCP_URL (public URL in the metadata), KALENDIA_API_URL (the Kalendia API).

Remaining for a public launch (tracked as Phase 2): hosting/deploy of this HTTP server, and full OAuth authorization-server discovery (dynamic client registration + auth-code + PKCE) so clients can mint tokens in-flow instead of pasting a kld_ token. Today a client connects by presenting a kld_ bearer (minted in Settings > API tokens).

Develop

uv run ruff check . && uv run ruff format --check . && uv run pyright && uv run pytest

Security

A token has the user's full account access (no scopes yet). Treat it like a password. Revoke a leaked or stale token in Kalendia: Settings > API tokens.

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选