Tempo MCP Server
Enables time tracking in Tempo with automatic timesheet generation from git commits, worklog management with roles and accounts, and integration with Jira for seamless project time logging.
README
Tempo MCP Server
MCP (Model Context Protocol) server for Tempo time tracking with full support for roles and accounts.
Features
- Auto-generate timesheets from git commits - scans all your repos and creates worklogs automatically
- Create worklogs with role and account attributes
- Automatically fetch account from Jira issue
- Auto-fallback to active account when issue's account is closed/archived
- Quick sprint meeting logging - just specify project and minutes
- Update and delete worklogs
- List worklogs for date ranges
- Get available work attributes and roles
Installation
npm install
npm run build
Configuration
Getting your API tokens
Tempo API Token:
- Go to Tempo > Settings in your Jira instance
- Click on API Integration in the left sidebar
- Click New Token
- Give it a name (e.g., "Claude Code MCP")
- Select permissions: Manage Worklogs (or Full Access)
- Copy the generated token
Jira API Token:
- Go to Atlassian API Tokens
- Click Create API token
- Give it a label (e.g., "Claude Code MCP")
- Copy the generated token
Environment Variables
| Variable | Required | Description |
|---|---|---|
TEMPO_API_TOKEN |
Yes | Tempo API token (get from Tempo > Settings > API Integration) |
JIRA_API_TOKEN |
Yes | Jira API token (get from id.atlassian.com/manage-profile/security/api-tokens) |
JIRA_EMAIL |
Yes | Jira account email |
JIRA_BASE_URL |
Yes | Jira base URL (e.g., https://company.atlassian.net) |
JIRA_ACCOUNT_FIELD_ID |
No | Custom field ID for Tempo account (default: 10026) |
DEFAULT_ROLE |
No | Default role for worklogs (default: Dev) |
DEFAULT_PROJECTS_DIR |
No | Default directory for git repos (e.g., C:/Users/you/Projects) |
DEFAULT_MONDAY_MEETING_ISSUE |
No | Jira issue for Monday team meeting (default: BS-14) |
Note:
gitAuthordefaults toJIRA_EMAILsince they're the same for all developers.
Usage with Claude Code
Quick Setup (one command)
Linux/macOS:
claude mcp add tempo-mcp -s user \
-e TEMPO_API_TOKEN=your-tempo-token \
-e JIRA_API_TOKEN=your-jira-token \
-e JIRA_EMAIL=your-email@company.com \
-e JIRA_BASE_URL=https://company.atlassian.net \
-e DEFAULT_PROJECTS_DIR=/home/you/projects \
-- npx -y github:Henry-Workshop/tempo-mcp
Windows (PowerShell):
claude mcp add tempo-mcp -s user `
-e TEMPO_API_TOKEN=your-tempo-token `
-e JIRA_API_TOKEN=your-jira-token `
-e JIRA_EMAIL=your-email@company.com `
-e JIRA_BASE_URL=https://company.atlassian.net `
-e DEFAULT_PROJECTS_DIR=C:/Users/you/Projects `
-- npx.cmd -y github:Henry-Workshop/tempo-mcp
Example configuration in claude.json
{
"mcpServers": {
"tempo-mcp": {
"command": "npx",
"args": ["-y", "github:Henry-Workshop/tempo-mcp"],
"env": {
"TEMPO_API_TOKEN": "your-tempo-token",
"JIRA_API_TOKEN": "your-jira-token",
"JIRA_EMAIL": "your-email@company.com",
"JIRA_BASE_URL": "https://company.atlassian.net",
"DEFAULT_PROJECTS_DIR": "C:/Users/you/Projects"
}
}
}
}
Available Tools
tempo_generate_timesheet
The main feature! Automatically generate your weekly timesheet from git commits.
{
"weekStart": "2025-12-01",
"gitAuthor": "your-email@company.com",
"projectsDir": "C:/Users/you/Projects",
"dryRun": true,
"mondayMeetingIssue": "BS-14"
}
| Parameter | Required | Description |
|---|---|---|
weekStart |
Yes | Monday date of the week (YYYY-MM-DD) |
gitAuthor |
Yes | Your git author email or name |
projectsDir |
Yes | Directory containing all your git repositories |
dryRun |
No | If true, shows plan without creating worklogs (default: false) |
mondayMeetingIssue |
No | Jira issue for Monday team meeting (default: BS-14) |
What it does:
- Scans all git repos in
projectsDir - Extracts your commits for Monday-Thursday
- Parses Jira issue keys from commit messages (e.g.,
PROJ-123) - Distributes 8h/day based on commit count per issue
- Adds 15min daily sprint meeting (on the main project of the day)
- Adds 15min Monday team sync on BS-14
- Generates client-friendly descriptions from commit messages
- Creates worklogs in Tempo
Example usage in Claude Code:
- "fait ma feuille de temps de cette semaine"
- "generate my timesheet for last week"
- "do my timesheet for the week of december 1st in dry run mode"
Important: Your commit messages should include Jira issue keys (e.g., PROJ-123 fix: description). Commits without issue keys will be flagged as warnings.
tempo_log_sprint_meeting
Quick way to log time for sprint meetings (daily, planning, retro). Automatically finds the "Sprint Meetings" issue for the project.
{
"projectKey": "PROJ",
"timeSpentMinutes": 15,
"description": "Daily",
"date": "2025-11-27"
}
Example usage in Claude Code:
- "add 15 mins for sprint meetings for BERGA"
- "log daily 15 mins to CRC project"
tempo_create_worklog
Create a new worklog with role and account support.
{
"issueKey": "PROJ-123",
"timeSpentHours": 1.5,
"date": "2025-11-27",
"description": "Working on feature",
"role": "Dev",
"accountKey": "ACCOUNT-KEY"
}
If accountKey is not provided, it will be automatically fetched from the Jira issue. If the account is closed/archived, it will automatically find an active account from recent project issues.
tempo_get_worklogs
Retrieve worklogs for a date range.
{
"startDate": "2025-11-01",
"endDate": "2025-11-30"
}
tempo_update_worklog
Update an existing worklog.
{
"worklogId": "12345",
"timeSpentHours": 2,
"description": "Updated description"
}
tempo_delete_worklog
Delete a worklog.
{
"worklogId": "12345"
}
tempo_get_work_attributes
Get available work attributes (roles, accounts configuration).
tempo_get_roles
Get available Tempo roles.
License
MIT
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。