This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
- Project Overview: See README.md for comprehensive information about CDviz architecture, components, and technologies
- Development Setup: See CONTRIBUTING.md for detailed development commands, workflow, and component-specific instructions
- Architecture Details: See README.md for event flow, database design, dashboard system, and deployment options
All development commands use mise task runner. Each component has a .mise.toml file with specific tasks:
cd {{component}}
mise install # Install component dependencies
mise tasks # List available tasks
mise run ci # Run all CI tasks (build, test, lint)mise run dev- Start development servermise run build- Build documentation
mise run build- Generate dashboards from TypeScriptmise run test- Run dashboard tests
mise run plan- Create migration for schema changesmise run apply- Apply migrationsmise run db-local:start- Start local database
mise run lint:*- Lint Helm chartsmise run publish- Publish to OCI registry
mise run stack:compose:up- Start Docker Compose stackmise run stack:k8s:deploy- Deploy to Kubernetesmise run use_cases:run- Run with sample data
CDviz is an SDLC observability platform using:
- Event Flow: Sources → cdviz-collector → PostgreSQL → Grafana dashboards
- Database: PostgreSQL + TimescaleDB with Atlas migrations
- Dashboards: Generated TypeScript using Grafana Foundation SDK with custom D3.js panels
- Deployment: Docker Compose for local dev, Helm charts for Kubernetes
- Direct Database Access: Dashboards query PostgreSQL directly (not through API)
- Generated Dashboards: Modify TypeScript in
dashboards_generator/, not JSON directly - Schema Changes: Use Atlas migrations via
mise run planin cdviz-db/ - Multi-Component: Each directory has separate
.mise.tomlconfiguration - CDEvents Standard: All events follow Cloud Native Delivery Events specification
For complete details, refer to README.md and CONTRIBUTING.md.