🐞 AI Bug Reporter

A Streamlit-based tool for generating detailed bug reports using OpenAI and Google Gemini models, with automatic issue creation in Jira.

---

Features

• Free-text bug description with optional screenshot upload (BLIP-based captioning).
• AI-powered JSON bug formatting: title, description, steps to reproduce, expected result, and actual result.
• Multi-provider support: Choose between OpenAI (gpt-4o-mini, gpt-4.1-nano, etc.) and Google Gemini (gemini-2.5-flash, gemini-2.5-pro, etc.).
• Interactive UI: fully editable bug reports before submission.
• Jira integration:
  • Configure via environment variables or sidebar input.
  • Dropdowns for Environment and Fix Versions.
  • Automatic ADF payload creation and issue submission.
• Loading spinners for all long-running operations (image processing, AI generation, Jira API calls).
• Modular codebase: easy maintenance and extension (ai_models.py, jira_utils.py, etc.).
• Gemini Model Discovery: Includes a utility script find_available_models.py to list all Google Gemini models accessible to your API key.

---

Prerequisites

• Python 3.10+
• An OpenAI API key (get from OpenAI)
• A Google Gemini API key (get from Google AI Studio)
• A Jira Cloud instance with:
  • A custom Environment single-select field (e.g., QA, Production, Staging, etc.)
  • Configured Fix Versions in your project
  • An API token from Atlassian API tokens

---

⚠️ Jira Custom Fields Mapping

Important: To enable issue creation in Jira, you must update the custom field IDs in settings.py (JIRA_CUSTOMFIELDS) to match your own Jira project’s configuration.

Example:

JIRA_CUSTOMFIELDS = {
    "steps":    "customfield_10420",
    "expected": "customfield_10422",
    "actual":   "customfield_10421",
    "env":      "customfield_10618"
}

• Find the custom field IDs in your Jira instance (Admin → Issues → Custom fields, or by inspecting the API).
• If the IDs don’t match, Jira will return field errors or the issue won’t be created as expected.

---

Installation

1. Clone this repository:

   git clone https://github.com/keinar/AI-Bug-Reporter.git
   cd AI-Bug-Reporter

2. Create a virtual environment and install dependencies:

   python3 -m venv venv
   source venv/bin/activate
   pip install --upgrade pip
   pip install -r requirements.txt

3. Create a .env file in the project root with the following variables (add only what you use):

   OPENAI_API_KEY=sk-...
   GEMINI_API_KEY=your-gemini-api-key
   JIRA_URL=https://your-domain.atlassian.net
   JIRA_EMAIL=you@company.com
   JIRA_API_TOKEN=your_jira_api_token
   JIRA_PROJECT_KEY=ABC

---

Running Locally

streamlit run ai_bug_reporter.py

Then open the URL shown in the terminal (e.g., http://localhost:8501).

---

Docker

1. Build the image:

   docker build -t ai-bug-reporter:latest .

2. Run the container (with your .env):

   docker run -p 8501:8501 --env-file .env ai-bug-reporter:latest

---

Usage

1. Enter a bug description in free text.
2. (Optional) Upload a screenshot for additional context.
3. Select Environment from the dropdown.
4. Select one or more Fix Versions.
5. Select your preferred AI model (OpenAI or Gemini).
6. Click Generate Bug to let the AI format the report.
7. Edit the generated bug details as needed.
8. Click Create in JIRA to submit the issue.
   • You’ll get immediate success/error feedback with a link to the new Jira issue.

---

Project Structure

.
├── ai_bug_reporter.py      # Main Streamlit application (UI)
├── ai_models.py            # OpenAI & Gemini API integration, parsing
├── jira_utils.py           # Jira payload building and API calls
├── image_captioning.py     # BLIP-based screenshot captioning
├── settings.py             # Constants & field mappings
├── requirements.txt        # Python dependencies
├── Dockerfile              # Container definition
├── .env                    # Environment variable definitions (local)
├── README.md               # This file
└── find_available_models.py# List available Gemini models for your key

---

Google Gemini Notes

• Get your API key from Google AI Studio.

• Not all Gemini models may be available to every account.
  To see which models you can use, run the included script:

  python find_available_models.py

• Use the full model name as shown (e.g., "models/gemini-2.5-flash").

---

Contributing

Pull requests welcome!
For major changes, please open an issue first to discuss what you would like to change.

---

License

MIT