jiragen 0.3.0

RAG-based JIRA ticket generator using vector similarity search

JiraGen: Automated JIRA Issue Generation

🚀 Overview

Jiragen is a CLI designed to automate the creation of JIRA Issues through the use of Local Large Language Models (LLMs). It leverages the power of Ollama and LiteLLM to provide context-aware issue generation, enabling efficient and effective interaction with JIRA. Full documentation is available here.

---

📖 Table of Contents

• Key Features
• Quick Start
• Configuration Options
• Template Customization
• Contributing
• License

🌟 Key Features

• 🧠 Local LLM Integration: Leverages Ollama and LiteLLM.
• 🔍 Context-Aware Issues: Incorporates relevant context using Vector store integration for scalable context handling.
• ✨ Customizable Templates: Adapt issue formats to your needs.
• 🔧 Metadata Extraction: Automates technical details from codebases.
• ⚙️ Configurable Parameters: Fine-tune the generation process.

⚡ Quick Start

Installation

Install JiraGen and its dependencies:

pip install jiragen

Install & run Ollama to use your local LLM:

curl https://ollama.ai/install.sh | sh
ollama pull phi4  # Replace with your preferred model

or export your OpenAI API key:

export OPENAI_API_KEY="YOUR_API_KEY"

Command Line Interface

JiraGen provides a powerful CLI for efficient ticket generation and management:

# Init
jiragen init

# Basic ticket generation
jiragen generate "Implement user authentication system"

# Generate with specific template
jiragen generate -t templates/custom.txt "Add API rate limiting"

# Generate with custom LLM configuration
jiragen generate --model codellama --temperature 0.8 "Refactor database schema"

# Status and management commands
jiragen status                    # Display vector store status
jiragen status --compact          # Show compact status view
jiragen status --depth 2          # Limit directory tree depth

# Vector store operations
jiragen add path/to/files         # Add files to vector store
jiragen rm path/to/files      # Remove files from vector store

API Usage

from jiragen import TicketGenerator, LLMConfig, GeneratorConfig
from pathlib import Path

# Configure the generator
config = GeneratorConfig(
    template_path=Path("templates/jira_template.txt"),
    llm_config=LLMConfig(model="llama2", api_base="http://localhost:11434"),
)

# Initialize the generator
generator = TicketGenerator(vector_store_client, config)

# Generate a ticket
ticket = generator.generate("Implement user authentication using JWT")

⚙️ Configuration Options

JiraGen supports a variety of configuration parameters to tailor ticket generation:

LLMConfig(
    model="llama2",  # Ollama model to use
    api_base="http://localhost:11434",  # Ollama endpoint
    max_tokens=2000,
    temperature=0.7,
    top_p=0.95,
)

📝 Template Customization

Create templates to match your organization's needs:

Title: {title}
Type: {type}
Priority: {priority}

Description:
{description}

Acceptance Criteria:
{acceptance_criteria}

Technical Implementation:
{implementation_details}

🤝 Contributing

We ❤️ contributions! To contribute:

1. Fork the repository.
2. Create a feature branch.
3. Submit a pull request with a detailed description of your changes.

For more details, refer to our CONTRIBUTING.md.

📜 License

JiraGen is released under the MIT License.