What's Next

Where BridgePort is heading, straight from theGitHub milestones — grouped into epics, in rough order of intent. Once something ships it moves to the changelog and drops off this page.

Directions, not commitments

Milestones describe intent and grouping, not dates or promises. Scope moves as reality does — the issue tracker is always the source of truth.

Milestone 4.0Up next

Progressive delivery & safe rollouts

0/3 done

Make multi-step rollouts (blue/green, canary, rolling, migration windows) a first-class, health-gated, revertible BridgePort operation — and round out the rollback story so any deploy can be undone fast.

Issues in this epic

• ○#321 Deploy: one-click rollback to any prior image version from deployment history
• ○#316 Deployment plans: phased rollouts with config/var/secret steps & ephemeral overrides (blue/green, canary, read-only migrations)
  • ○#338 Deployment plans ①: step-type engine + ephemeral overrides (set_config / set_var / set_secret / run_command / wait / approval / revert)
  • ○#339 Deployment plans ②: subset/tag targeting for deploy & config steps (canary/rolling enabler)
  • ○#340 Deployment plans ③: plan-builder UI + choreography dry-run visualization
• ○#317 Slack: thread deployment-plan notifications (single parent message + step replies + in-place status update)

Proactive alerting & observability

0/4 done

Close the "collects rich metrics but only alerts after something is already down" gap: threshold rules that fire *before* outages, notification of available image updates, more delivery channels, and a Prometheus endpoint for teams that already run Grafana.

Issues in this epic

• ○#318 Monitoring: metric threshold alerting (CPU/mem/disk/load + crash-loop/OOM rules → notifications)
• ○#322 Notifications: image-update-available notification (new digest detected)
• ○#325 Monitoring: Prometheus/OpenMetrics export endpoint (Grafana datasource)
• ○#326 Notifications: additional channels (Discord, Telegram, PagerDuty/Opsgenie; Apprise/Shoutrrr-style adapter)

Backup, restore & disaster recovery

0/2 done

Turn BridgePort's write-only backup story into a full DR loop: restore what it backs up, and back up BridgePort itself.

Issues in this epic

• ○#319 Databases: backup restore (one-click restore from local or S3)
• ○#328 Backups: control-plane (BridgePort instance) backup & restore / migration

Identity, access & compliance

0/5 done

The security wave: bring-your-own identity provider, a second factor for local accounts, secrets sourced from external stores, and audit logs that can leave the box — the four gaps that gate adoption by security-conscious teams.

Issues in this epic

• ○#324 Auth: two-factor authentication (TOTP) for local accounts
• ○#330 Security: audit-log export & streaming to external sink (CSV/JSON, syslog/SIEM)
• ○#323 Auth: OAuth/OIDC SSO login (bring-your-own IdP) wired into RBAC
• ○#109 Secrets: External secret providers (Vault, AWS, Passbolt)
• ○#354 Security: per-account login throttling + remove login user-enumeration timing

Service resources & structured configuration

0/3 done

Give the Service model typed, managed fields for the things operators currently hand-write into compose YAML — resource limits and volumes first, then the structured editor that surfaces all of it.

Issues in this epic

• ○#320 Services: per-service CPU/memory resource limits applied at deploy
• ○#327 Storage: first-class Docker volume & bind-mount management
• ○#108 Services: Structured compose editor + parsed view in UI

Milestone 5.0Later

Day-2 operations & automation

0/4 done

The 5.0 ops wave: run things (recurring jobs, interactive shells) and connect things (CI/CD webhooks, scripting) through BRIDGEPORT instead of around it. All four are fully specced (2026-07-04 design interview — see each issue's spec comment).

Issues in this epic

• ○#347 CLI: machine-readable JSON output (--output json) for scripting
• ○#346 Operations: scheduled commands / cron jobs in containers (recurring, with history & failure alerts)
• ○#348 CI/CD: native webhook receivers for GitHub / GitLab / Gitea / Harbor / Quay
• ○#329 Operations: browser-based web terminal to servers & containers

• ○#111 Ansible playbook orchestration [deferred from 2.0]
• ○#352 Security: hash agent tokens at rest
• ○#353 Security: explicit per-route role gates on services.ts & registries.ts (defense-in-depth)
• ○#354 Security: per-account login throttling + remove login user-enumeration timing
• ○#355 Epic: Trivy security scanning
• ○#357 SBOM export endpoint per container image
• ○#358 Misconfiguration scanning for compose/config files (trivy config)
• ○#359 Host OS package CVE scanning via the agent
• ○#360 License reporting for container image packages
• ○#361 CVE scanning & alerting for container images (bundled Trivy)
• ○#363 Backups: volume & app-data backup/restore (non-database)
• ○#364 Notifications: Opsgenie + URL-adapter long tail (ntfy, Gotify, Teams, Mattermost)

Want to help?

These are scoped and ready for outside contributions — comment on an issue to claim it.

Good first issues

• ○#353 Security: explicit per-route role gates on services.ts & registries.ts (defense-in-depth)
• ○#347 CLI: machine-readable JSON output (--output json) for scripting
• ○#322 Notifications: image-update-available notification (new digest detected)

Help wanted

• ○#354 Security: per-account login throttling + remove login user-enumeration timing
• ○#348 CI/CD: native webhook receivers for GitHub / GitLab / Gitea / Harbor / Quay
• ○#347 CLI: machine-readable JSON output (--output json) for scripting
• ○#346 Operations: scheduled commands / cron jobs in containers (recurring, with history & failure alerts)

Generated from the GitHub issue tracker when this site was last deployed (2026-07-04).