timetastic-mcp
An unofficial MCP server that exposes the Timetastic API to enable querying and managing absence & leave data.
README
Timetastic MCP
An unofficial MCP server for Timetastic, exposing the Timetastic API to agentic clients so they can query and manage absence & leave data.
Setup
You need an admin API token, generated at
https://app.timetastic.co.uk/api. The server reads it from the
TIMETASTIC_API_TOKEN environment variable.
The quickest way to run it is with uv's uvx,
which fetches and runs the published package without a manual install:
export TIMETASTIC_API_TOKEN="your-token-here"
uvx timetastic-mcp
Or install it as a tool so the timetastic-mcp command is on your PATH:
uv tool install timetastic-mcp # or: pipx install timetastic-mcp
[!IMPORTANT] Only admin users can generate API tokens on Timetastic. The above URL is only accessible to admin users.
Client configuration
Add the server to your MCP client (e.g. Claude Desktop / Claude Code):
{
"mcpServers": {
"timetastic": {
"command": "uvx",
"args": ["timetastic-mcp"],
"env": { "TIMETASTIC_API_TOKEN": "your-token-here" }
}
}
}
Pin a version with "args": ["timetastic-mcp@0.1.0"]. If you installed the
command on your PATH instead, use "command": "timetastic-mcp" with
"args": []. In Claude Code you can also add it from the CLI:
claude mcp add timetastic --env TIMETASTIC_API_TOKEN=your-token -- uvx timetastic-mcp
Tools
Tools are grouped by resource and named <verb>_<resource>.
| Group | Tools |
|---|---|
| Absences | list_absences |
| Holidays (leave bookings) | list_holidays, get_holiday, book_holiday, action_holiday |
| Users | list_users, get_user, get_user_contact, add_user, edit_user, archive_user, restore_user, assign_public_holidays_to_user |
| Departments | list_departments, get_department, add_department, edit_department, delete_department |
| Leave types | list_leave_types, get_leave_type, list_leave_type_colors, list_leave_type_icons, create_leave_type, update_leave_type, delete_leave_type |
| Allowances | list_all_allowances, get_user_allowance, update_user_allowance, update_user_carry_forward, add_user_toil, update_user_toil, delete_user_toil |
| Locked dates | list_locked_dates, add_locked_date, delete_locked_date |
| Public holidays | list_public_holidays, get_public_holiday, list_public_holiday_countries |
| Webhooks | list_webhook_events |
Note: Timetastic calls all leave bookings "holidays" for historical reasons — the Holidays tools cover any kind of absence, not just annual leave.
Layout
The package lives under src/timetastic_mcp/:
timetastic.py— async HTTP client (auth, base URL, rate-limit retries, error handling).server.py— the sharedFastMCPinstance andget_client(), the lazily-created API client the tools call.tools/— one module per resource area, each registering its tools on the shared server:absences,holidays,users,departments,leave_types,allowances,locked_dates,public_holidays,webhooks.main.py— entry point (thetimetastic-mcpconsole script); importstoolsto register everything, then runs the server.
Development
Requires Python 3.13+ and uv.
uv sync # create the venv and install the package + dev deps
uv run pytest # run the test suite
uv run timetastic-mcp # run the server from your checkout
Notes
- The API is rate limited to 5 requests/second per token (1/second for
list_absences); the client retries once on a429. - Write and admin operations require appropriate permissions on the token's user account.
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。