Deployment & White-Label

Deployment choices that pass review

ATK supports three delivery patterns that run the same code and APIs:

Procurement sees real controls instead of promises: you can point to the charts, the pipelines, and the evidence before diligence even starts.

Branding without rewrites

• The Next.js app (kit/dapp) ships theming, tokens, and brand packs so tenants can match their visual identity without forking screens.
• Helm ingress values let you attach custom domains, certificates, and CDN strategies to each surface while keeping TLS under your policy.
• Portal APIs and webhooks stay consistent, letting partners embed flows or build new UIs while regulated functions remain in your environment.

Infrastructure building blocks

• Hyperledger Besu, Blockscout, and supporting services swap through values overrides; images, storage classes, or cloud primitives change without touching templates.
• Tooling such as kit/charts/tools/aws-marketplace-automation.ts and the environment hierarchy document promotion paths from local to staging to production, including feature-branch environments.
• Observability is bundled: Prometheus, Grafana, Loki, and alert routing stream metrics and logs into your SIEM from day one.

Security, identity, audit

• Secrets land in Kubernetes or external stores, and charts expose OIDC hooks for services like MinIO while network policies enforce least privilege between pods.
• The dApp and REST API backend (documented via OpenAPI and implemented in the orpc module) already model OAuth flows, MFA checks, and granular role validation, so existing identity providers plug straight in.
• Every component emits structured logs and metrics; forwarding them to your observability stack gives compliance the audit trails it expects without extra engineering.

Operational readiness

• bun run ci packages formatting, contract compilation, code generation, linting, and testing; the same workflow underpins Console deployments and self-managed pipelines.
• Deployment docs cover backups, disaster recovery, and capacity planning, cutting the time needed to answer RTO/RPO questions.
• Moving between models is configuration work: Compose for development, Helm for self-managed clusters, and Console for managed service all share contracts, schema, and the same OpenAPI-driven REST endpoints.

Bottom line: ATK is deployable market infrastructure. Own the runtime, match your brand, and let every other chapter build on a stack you already control.