Changelog¶
All notable changes to NextSight AI are documented here.
[1.4.0] - 2025-12-17¶
Added¶
AI-Powered Features¶
- Proactive Insights -
/api/v1/ai/insights/proactiveendpoint for automatic issue detection - Failed/crashing pod detection
- Degraded deployment identification
- Security risk highlighting
- Resource optimization opportunities
- Cluster health scoring
- AI Runbook Generation -
/api/v1/ai/runbook/generateendpoint for incident response - Support for pod_crash, deployment_failed, node_not_ready, oom_killed, network_issue
- Step-by-step kubectl commands with expected outputs
- Verification and escalation paths
- Smart Suggestions -
/api/v1/ai/suggestions/smartendpoint for context-aware recommendations - Real-time cluster state analysis
- Actionable kubectl commands
Security & Monitoring¶
- Trivy in Docker Image - Built-in vulnerability scanning
- Trivy v0.58.0 binary included in backend image
- No external dependencies required
- Works in air-gapped environments
- Metrics Server Status Endpoint -
/api/v1/kubernetes/metrics/status - Auto-detect metrics-server availability
- Graceful fallback when unavailable
- Top 3 Security Risks View - Priority security findings on dashboard
- Plain English Security Explanations - Business-friendly risk messaging with impact assessment
User Experience Enhancements¶
- Dashboard Actionable Insights Card - AI-powered recommendations on main dashboard
- Workload Health Indicators - Visual health status (green/yellow/red) for all workloads
- WorkloadDrawer AI Fixes Tab - Root cause analysis and intelligent remediation
- YAML Deploy Summary Modal - Pre-deployment overview and validation
Changed¶
- Project Rebrand to NextSight AI
- New name and tagline: "See your DevOps world in one intelligent view"
- Updated all UI components, configs, and documentation
- Helm chart description updated
Improved¶
- Enhanced Redis Caching - Production-ready caching with optimized TTLs
- Optimization dashboard: 60s TTL
- Prometheus targets: 30s TTL
- Prometheus alerts: 15s TTL
- AI proactive insights: 120s TTL
- Production Readiness
- Comprehensive environment configuration
- Production checklist in
.env.example - Proper error handling across all services
- Graceful degradation when services unavailable
Deferred to Future Release¶
- Pipelines Module - Full CI/CD pipeline management with agent-based execution
- Cost Analyzer - Resource cost tracking, namespace breakdown, and savings recommendations
[1.3.1] - 2024-12-05¶
Fixed¶
- CodeQL security alerts (log injection)
- ESLint @typescript-eslint version conflict
- Empty except blocks in health checks
Changed¶
- Removed OSSF Scorecard (permission conflicts)
- Updated dependencies (date-fns, websockets, redis)
[1.3.0] - 2024-12-04¶
Added¶
- Pod Exec terminal with full PTY support
- Debug containers for distroless images
- WebSocket-based log streaming
- Terminal resize support
Fixed¶
- XSS vulnerability in log viewer
- Stack trace exposure in AI health check
[1.2.0] - 2024-12-03¶
Added¶
- Security Posture Dashboard
- Trivy vulnerability scanning
- RBAC analysis
- Network policy coverage
- AI-powered remediation suggestions
Changed¶
- Improved security score algorithm
- Better finding categorization
[1.1.0] - 2024-12-02¶
Added¶
- AI Chat Assistant (Gemini integration)
- Real-time Kubernetes data in AI responses
- Dark mode support
Changed¶
- Updated UI with glass-morphism design
- Improved response times
[1.0.0] - 2024-12-01¶
Added¶
- Initial release
- Kubernetes dashboard
- Pod/Deployment/Service management
- Node monitoring
- kubectl terminal
- Multi-cluster support
- RBAC authentication
Version Format¶
We follow Semantic Versioning:
- MAJOR - Breaking changes
- MINOR - New features (backwards compatible)
- PATCH - Bug fixes
Contributing¶
See Contributing Guide for how to contribute.