TrackingTime MCP Server
Connects AI assistants to the TrackingTime API v4 for managing projects, tasks, and team assignments. Users can start or stop timers, log manual time entries, and organize project workflows using natural language.
README
TrackingTime MCP Server
An MCP (Model Context Protocol) server that connects AI assistants like Claude to the TrackingTime API v4. Manage projects, tasks, time tracking, staff assignments, and customers through natural language.
Quick Start
1. Get your TrackingTime credentials
- App Password: TrackingTime → Manage → User Settings → Apps & Integrations → create a new App Password
- Account ID: Visible in your TrackingTime URL when logged in, or in account settings
2. Add to your AI assistant
Claude Code:
claude mcp add trackingtime -e TT_APP_PASSWORD=your-app-password -e TT_ACCOUNT_ID=your-account-id -- npx trackingtime-mcp
Claude Desktop (claude_desktop_config.json):
{
"mcpServers": {
"trackingtime": {
"command": "npx",
"args": ["trackingtime-mcp"],
"env": {
"TT_APP_PASSWORD": "your-app-password",
"TT_ACCOUNT_ID": "your-account-id"
}
}
}
}
Restart your assistant after adding.
Alternative: Install from source
If you prefer to clone and build locally:
git clone https://github.com/ficus33/trackingtime-mcp.git
cd trackingtime-mcp
npm install
npm run build
cp .env.example .env # then edit .env with your credentials
Then point your assistant at the local build:
claude mcp add trackingtime -- node /path/to/trackingtime-mcp/dist/index.js
Tools
Projects
| Tool | Description |
|---|---|
tt_list_projects |
List projects (filter: ACTIVE/ARCHIVED/ALL/FOLLOWING) |
tt_list_project_ids |
List only project IDs (lightweight, for batch operations) |
tt_search_projects |
Search projects and tasks by keyword |
tt_create_project |
Create a new project |
tt_update_project |
Edit project name, customer, or service |
tt_update_project_preferences |
Set favorite, default view, show closed tasks |
tt_get_project |
Get single project with detail flags |
tt_get_project_times |
Get accumulated time for multiple projects |
tt_get_project_users |
See which staff are on a project |
tt_archive_project |
Archive a project (reversible) |
tt_reopen_project |
Reopen an archived project |
tt_delete_project |
Permanently delete a project |
tt_merge_projects |
Merge one project into another |
Tasks
| Tool | Description |
|---|---|
tt_list_tasks |
List tasks (filter: ACTIVE/ARCHIVED/ALL/TRACKING) |
tt_create_task |
Create a task with assignees, due date, estimate |
tt_update_task |
Edit task or reassign staff |
tt_get_task |
Get single task details |
tt_get_task_times |
Get accumulated times for multiple tasks |
tt_search_tasks |
Search tasks by name within projects |
tt_sort_tasks |
Reorder tasks by sort index |
tt_import_tasks |
Bulk import tasks with preview mode |
tt_close_task |
Mark a task as complete |
tt_reopen_task |
Reopen a completed task |
tt_delete_task |
Delete a task |
Time Tracking
| Tool | Description |
|---|---|
tt_start_timer |
Start a timer on a task |
tt_stop_timer |
Stop a running timer |
Time Entries
| Tool | Description |
|---|---|
tt_list_time_entries |
List entries by user/project/customer/task + date range |
tt_add_time_entry |
Add a manual time entry (duration in seconds) |
tt_get_time_entry |
Get a single time entry |
tt_update_time_entry |
Edit a time entry |
tt_delete_time_entry |
Delete a time entry |
tt_get_events_summary |
Summary of hours per user per day |
tt_export_time_entries |
Export as CSV |
tt_mark_billed |
Flag entries as billed |
tt_mark_not_billed |
Unflag billed entries |
Users & Staff
| Tool | Description |
|---|---|
tt_list_users |
List all staff (find user IDs) |
tt_get_user |
Get single user details |
tt_update_user |
Update user profile |
tt_get_user_tasks |
List a user's tasks grouped by project |
tt_get_user_tracking |
See what a user is currently tracking |
tt_get_user_trackables |
All projects and tasks assigned to a user |
tt_get_user_projects |
List projects assigned to a user |
tt_assign_user_projects |
Assign staff to projects |
tt_remove_user_projects |
Remove staff from projects |
tt_archive_user |
Deactivate a user (admin only) |
tt_reactivate_user |
Reactivate an archived user |
tt_invite_users |
Invite people by email |
Customers
| Tool | Description |
|---|---|
tt_list_customers |
List customers (filter: ACTIVE/ARCHIVED/ALL) |
tt_get_customer |
Get single customer details |
tt_create_customer |
Create a new customer |
tt_update_customer |
Edit customer details |
tt_delete_customer |
Permanently delete a customer |
tt_archive_customer |
Archive a customer (reversible) |
tt_reactivate_customer |
Reactivate an archived customer |
Reducing tool count
All 54 tools are active by default. If you find this adds too much context for your AI assistant, you can clone the repo, comment out tools you don't need in src/tools.ts with /* */, and run npm run build to create a slimmer build.
Testing
Use MCP Inspector to test tools interactively:
npx @modelcontextprotocol/inspector npx trackingtime-mcp
API Notes
A few TrackingTime API quirks to be aware of:
durationandaccumulated_timeare in secondsestimated_timeandworked_hoursare in hours- Time entries are called "events" in the API
- Starting a timer when one is already running returns error 502 — use
stop_running_task=trueto auto-stop the current timer - Dates use
YYYY-MM-DD, datetimes useyyyy-MM-dd HH:mm:ss
Auth
This server uses TrackingTime's App Password authentication. Your real password is never stored. The App Password is sent as HTTP Basic auth (API_TOKEN:<app_password>) over SSL.
If an App Password is compromised, revoke it in TrackingTime and create a new one — no need to change your account password.
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。