English | Русский

JT - Jira Time Tracker

AI-powered Jira time logging CLI. Write in natural language and confirm the result before it reaches Jira.

Installation

npm install -g jtw

Quick Start

# First-time setup
jtw setup

# Quick AI logging
jtw q "today PROJ-123 development 2h"

# Quick AI logging with alias-style work
jtw q "today standup 30m"

# Interactive mode
jtw

# Manage templates
jtw t

# Manage aliases
jtw a

Features

• AI parsing: Natural language to structured worklogs with preview and confirmation
• Smarter fallbacks: If AI cannot resolve a task, JTW suggests recent tasks, history, and aliases
• Aliases: Map recurring work to Jira tasks, for example standup -> PROJ-123
• Templates: Save and reuse common worklog batches
• History: All successful logs are stored locally in SQLite for better suggestions
• Periods: last week standups 1.5h each day expands to workdays automatically
• Jira Server & Cloud: Works with on-premise and cloud installations

Commands

Command	Description
jtw	Interactive menu
jtw setup	Configure Jira, AI, and language settings
jtw q "text"	Parse a natural-language worklog and log it after confirmation
jtw t	Template management
jtw a	Alias management

Configuration

All settings are stored in ~/.jtw/data.db (SQLite). Run jtw setup to configure:

• Jira URL, username, password
• Project key, for example PROJ
• AI provider (Anthropic or OpenAI) and API key
• CLI language (English or Russian)

When you rerun jtw setup, saved values are prefilled. Leave secret fields empty to keep the existing password and API key.

Common Flows

First setup

jtw setup

After a successful connection test, JTW shows concrete next commands you can run immediately.

AI quick log

jtw q "today PROJ-123 code review 1h, standup 30m"

JTW parses the text, validates the Jira tasks, shows a grouped preview by date, and asks for confirmation before logging.

Resolve an unresolved task

jtw q "today incident follow-up 45m"

If AI cannot determine the Jira task, JTW offers:

• recent logged tasks from local history
• recent Jira issues assigned to you
• saved aliases
• manual task entry

Aliases and templates workflow

# Save recurring work keywords
jtw a

# Save reusable batches of entries
jtw t

Aliases help AI resolve recurring work. Templates help with repeated multi-entry logs.

Examples

# Simple log
jtw q "today standup 30 minutes"

# Multiple entries at once
jtw q "yesterday PROJ-123 development 6 hours, review 2 hours"

# AI matches aliases by meaning
jtw q "today standups 2 hours"

# Periods expand to workdays
jtw q "last week standups 1.5h each day"

# Relative dates
jtw q "day before yesterday bugfix 4h"

Requirements

• Node.js 18+
• Jira Server or Jira Cloud access
• Anthropic or OpenAI API key

Troubleshooting

Cannot connect to Jira: Run jtw setup again. The connection step now distinguishes credentials, network or VPN, and SSL certificate issues.

AI returned invalid output: Retry the command or simplify the wording. If the issue repeats, try an explicit task key like PROJ-123.

better-sqlite3 won't install: The package includes a native dependency. Prebuilt binaries are downloaded automatically. If that fails, you need a C++ compiler:

• macOS: xcode-select --install
• Ubuntu: sudo apt install build-essential python3
• Windows: npm install -g windows-build-tools

Architecture

src/
├── commands/      # CLI commands (setup, quick, template, alias, log)
├── core/          # Core logic (store, jira, ai)
├── utils/         # Utilities (display, menu)
├── types.ts       # TypeScript types
└── index.ts       # CLI entry point

Minimalist design, no service layers. SQLite stores configuration, aliases, templates, and log history.

License

MIT