Skip to content

Getting Started

This guide will help you get up and running with JiraGen.

Installation

First, install Jiragen using pip:

pip install jiragen

Initial Setup

Configuration

Run the initialization command to set up Jiragen:

jiragen init

This will: 1. Create the necessary directories following XDG Base Directory specification: - Configuration directory: - Linux/macOS: ~/.config/jiragen/ - Windows: %APPDATA%\jiragen\ - Data directory: - Linux/macOS: ~/.local/share/jiragen/ - Windows: %LOCALAPPDATA%\jiragen\ 2. Guide you through configuring: - JIRA connection settings - LLM preferences

You can accept the defaults by pressing Enter or customize each setting.

Environment Variables

For automated setup, you can use environment variables:

# JIRA Settings
export JIRAGEN_URL="https://your-domain.atlassian.net"
export JIRAGEN_USERNAME="your-email@example.com"
export JIRAGEN_API_TOKEN="your-api-token"
export JIRAGEN_DEFAULT_PROJECT="PROJECT"
export JIRAGEN_DEFAULT_ASSIGNEE="username"

# LLM Settings
export JIRAGEN_LLM_MODEL="gpt-4"
export JIRAGEN_LLM_TEMPERATURE="0.7"
export JIRAGEN_LLM_MAX_TOKENS="2000"

# Optional: Custom paths
export XDG_CONFIG_HOME="~/.config"      # Unix-like systems
export XDG_DATA_HOME="~/.local/share"   # Unix-like systems

Configuration

1. Initialize JiraGen

First, initialize JiraGen:

jiragen init

This will create the following directory structure following the XDG Base Directory specification:

Unix-like Systems (Linux, macOS): 

~/.config/jiragen/           # Configuration directory
├── config.ini              # Main configuration file
└── templates/             # Custom templates
    └── default.md

~/.local/share/jiragen/     # Data directory
├── vector_store/          # Vector database storage
├── jira_data/            # JIRA fetched data
│   ├── epics/           # Epic data
│   ├── tickets/         # Ticket data
│   └── components/      # Component data
└── jiragen.log          # Application logs

Windows: 

%APPDATA%\jiragen\          # Configuration directory
├── config.ini              # Main configuration file
└── templates/             # Custom templates
    └── default.md

%LOCALAPPDATA%\jiragen\     # Data directory
├── vector_store/          # Vector database storage
├── jira_data/            # JIRA fetched data
│   ├── epics/           # Epic data
│   ├── tickets/         # Ticket data
│   └── components/      # Component data
└── jiragen.log          # Application logs

2. Configure JIRA Settings

Edit the configuration file (~/.config/jiragen/config.ini on Unix-like systems or %APPDATA%\jiragen\config.ini on Windows):

[jira]
url = https://your-domain.atlassian.net
username = your-email@example.com
api_token = your-api-token
default_project = PROJECT
default_assignee = username

[llm]
model = gpt-4
temperature = 0.7
max_tokens = 2000

You can also use environment variables: 

export JIRA_URL=https://your-domain.atlassian.net
export JIRA_USERNAME=your-email@example.com
export JIRA_API_TOKEN=your-api-token

3. Add Your Codebase

Index your codebase to provide context for ticket generation:

jiragen add .

This will: 1. Scan your codebase recursively 2. Respect .gitignore patterns (skipping ignored files) 3. Create embeddings for relevant files 4. Store them in the vector store

You can also: - Add specific files: jiragen add src/main.py tests/test_api.py - Add a specific directory: jiragen add src/ - Add files matching a pattern: jiragen add *.py

The command will show: - Progress bar during processing - Tree view of added files - Processing statistics (files/second)

Example output: 

📁 Added Files
├── src/main.py
├── src/utils/helper.py
└── tests/test_api.py

Successfully added 3 files (52.7 files/second)

Basic Usage

Generate a Ticket

jiragen generate "Add dark mode support"

This will: 1. Search your staged jira data and codebase content for relevant context 2. Generate a detailed ticket description 3. Open it in your editor for review 4. Extract metadata (issue type, priority, labels)

Generate and Upload

jiragen generate "Add dark mode support" --upload

This will: 1. Generate the ticket content 2. Show extracted metadata for review 3. Upload to JIRA after confirmation

Automated Pipeline

jiragen generate "Add dark mode support" --upload --yes

This will: 1. Generate the ticket 2. Extract metadata 3. Upload to JIRA without confirmation

Templates

Default Template

JiraGen comes with a default template that structures tickets with: - Description - Acceptance Criteria - Technical Details - Implementation Notes

Custom Templates

Create custom templates in the templates directory: - Unix-like: ~/.config/jiragen/templates/ - Windows: %APPDATA%\jiragen\templates\

# {title}

## Description
{description}

## Implementation Plan
- [ ] Task 1
- [ ] Task 2

## Technical Requirements
{technical_details}

## Testing Strategy
- Unit Tests
- Integration Tests
- E2E Tests

Use your template: 

jiragen generate "Feature request" --template custom.md

Next Steps