A comprehensive tool for extracting and analyzing incident data from Jira and Confluence, with advanced RCA (Root Cause Analysis) content analysis capabilities.
This tool pulls incident data from Hinge Health's IR (Incident Recovery) board in Jira, matches incidents with their corresponding RCA documents in Confluence, and provides detailed analysis including:
- Incident Data Extraction: Complete incident metadata, urgency levels, and team involvement
- RCA Document Matching: Automatic linking to Confluence RCA documents
- Enhanced Analysis: LLM-powered extraction of incident summaries, user impact, and root causes
- CSV Export: Structured data export for further analysis and reporting
- Jira Integration: Extract all IR incidents since January 1, 2024
- Custom Field Support: Incident Urgency (P1-P4) and Pods Engaged
- Confluence Integration: Automatic RCA document discovery and linking
- Data Quality: 126 incidents retrieved, 72.2% RCA match rate
- CSV Export: Clean, structured data export
- LLM-Powered RCA Analysis: Extract structured insights from RCA documents
- Enhanced Data Fields: Incident summaries, user impact metrics, root cause analysis
- Quality Validation: Analysis confidence scoring and validation
- Advanced Reporting: Comprehensive data quality and analysis reports
- Python 3.8+
- Atlassian account with access to Hinge Health Jira and Confluence
- API token for Atlassian services
-
Clone the repository:
git clone https://github.com/yourusername/ir-analysis.git cd ir-analysis -
Install dependencies:
pip install -r requirements.txt
-
Configure environment:
cp config/.env.example config/.env # Edit config/.env with your Atlassian credentials -
Run analysis:
python src/main.py
Create config/.env with your Atlassian credentials:
JIRA_URL=https://hingehealth.atlassian.net
CONFLUENCE_URL=https://hingehealth.atlassian.net/wiki
ATLASSIAN_EMAIL=[email protected]
ATLASSIAN_API_TOKEN=your-api-token
LOG_LEVEL=INFOoutput/incidents_2024_*.csv: Main incident data exportlogs/incident_analysis_*.log: Detailed execution logsoutput/rca_analysis_summary.json: RCA analysis results (coming soon)
| Column | Description |
|---|---|
| Ticket Key | Jira ticket identifier (IR-XXX) |
| Summary | Incident title |
| Incident Urgency | Priority level (P1-P4) |
| Jira Description | Full incident description |
| RCA Link | URL to Confluence RCA document |
| Pods Engaged | Teams involved in incident response |
| Created | Incident creation timestamp |
| Status | Current ticket status |
| Column | Description |
|---|---|
| Incident Summary | LLM-generated executive summary |
| Users Impacted | Extracted user impact data |
| Root Causes | Structured root cause analysis |
| Analysis Quality | Confidence score of LLM analysis |
ir-analysis/
├── src/
│ ├── jira_client.py # Jira API integration
│ ├── confluence_client.py # Confluence API integration
│ ├── rca_analyzer.py # LLM-based RCA analysis (coming soon)
│ └── main.py # Main execution script
├── config/
│ ├── .env.example # Environment template
│ └── requirements.txt # Python dependencies
├── output/ # Generated reports
├── logs/ # Application logs
├── PLAN.md # Detailed project plan
└── README.md # This file
This project follows a phased development approach:
- Phase 1: Data verification and API connectivity ✅
- Phase 2: RCA document matching ✅
- Phase 3: Basic data extraction and CSV generation ✅
- Phase 4: LLM-powered RCA content analysis 🚧
- Phase 5: Advanced validation and reporting 🚧
See PLAN.md for detailed implementation roadmap.
- Incidents Retrieved: 126 (since Jan 1, 2024)
- RCA Match Rate: 72.2% (91/126 incidents)
- Execution Time: ~2-3 minutes for full analysis
- Success Rate: 100% for available data
- Create a feature branch from
main - Make your changes with appropriate tests
- Update documentation as needed
- Submit a pull request
Internal Hinge Health tool - not for external distribution.
For issues or questions, please create an issue in this repository or contact the development team.