Jira AI Agent 🤖

Natural language interface for Jira task management, powered by AI.

Features

🤖 Natural language task creation 🔒 Safety validation layers
🔄 Multiple LLM support (OpenAI, Ollama) 🌐 Web UI + CLI interfaces ✨ Real-time HTMX updates 🧪 Test coverage & mocking

Installation

1. Clone and setup:

git clone https://github.com/YOUR_USER/jira-ai-agent.git
cd jira-ai-agent
cp .env.example .env

2. Configure environment:

# Jira Config
JIRA_SERVER=https://your-domain.atlassian.net
JIRA_USER[email protected] 
JIRA_TOKEN=your-api-token

# LLM Provider
LLM_PROVIDER=openai  # or deepseek, llama
OPENAI_API_KEY=your-api-key
OPENAI_MODEL=gpt-3.5-turbo

Usage

Docker (Recommended)

# Start all services
docker compose up --build

# CLI only
docker compose run cli

Web UI: http://localhost:8000

Local Development

# Setup
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt

# Run
python src/web.py  # Web UI
python src/cli.py  # CLI

Testing

Run test suite:

./scripts/run-tests.sh

Configure test env:

# .env.test
JIRA_SERVER=http://mock-jira
JIRA_USER=test-user 
JIRA_TOKEN=test-token
LLM_PROVIDER=openai
OPENAI_API_KEY=test-api-key

Run specific tests:

pytest tests/test_integration.py::TestJiraIntegration::test_valid_command
pytest --cov=src tests/

Mock Configuration

Test mocks configured in

conftest.py

:

• OpenAI API responses
• Jira API calls
• Ollama API endpoints

License

MIT License - See LICENSE file for details