Release rollouts with rings, auto-advance, and rollback
Ship backend changes through canary percentages, health gates, staged promotion, and forced rollback when a release goes bad.
What it doesLets a release move through explicit rollout rings such as 5%, 25%, and 100%, with deterministic client bucketing, dwell-time checks, telemetry gates, and environment-aware deployment control.
Why it mattersMost backend platforms can deploy. Fewer can limit blast radius, auto-promote on health, and force a rollback to the last known good release when a bad build slips through.
Real-world useRoll a new white-label app backend to a small subscriber slice first, watch runtime success and error rates, then widen the rollout only if the system stays healthy.
Collapse+
Cross-product compatibility rules and advisory feeds
Model release dependencies, security advisories, and forced security upgrades directly in the platform.
Read more+
Public marketplace catalogue and private allowlist apps
Run a public app store when you want reach, then lock selected apps behind email allowlists when distribution must stay controlled.
Read more+
Metered add-ons, tier switching, and subscriber grants
Charge per seat, per screen, or per event, while still supporting flat add-ons, trials, grants, and mid-cycle tier changes.
Read more+
Passthrough Stripe payments with Connect commission splits
Charge through your own Stripe account, refund safely, audit every event, and split funds in marketplace flows without Vault56 holding the money.
Read more+
Organization AI wallet with per-member spend limits
Fund AI Builder at the org level, cap who can spend it, and keep a ledger of which members used what budget on which projects.
Read more+
Short-lived TURN credentials for WebRTC apps
Issue expiring TURN credentials for restrictive networks so realtime audio, video, and screen-share apps can relay safely.
Read more+
Atomic capability creation for app bundles and pricing
Create features, tier bundling, and optional paid add-ons in one transaction instead of stitching together fragile multi-call setup flows.
Read more+
Preview URLs with tier impersonation and usage caps
Generate tightly scoped preview sessions that behave like a subscriber on a chosen tier without exposing full publisher power.
Read more+
Sparse-field projection and high-speed batch inserts
Trim payloads with `_fields` and push large writes through bulk insert paths that preserve API shape while removing needless overhead.
Read more+
Live IMAP peek for inbox experiences without full sync
Read email directly from IMAP on demand, even when the account is not running the heavier background sync pipeline.
Read more+
App-token self-introspection and stricter preview scopes
Let a running app understand exactly which app token it is using, while keeping preview tokens locked to a narrow allowlist of safe paths.
Read more+
Draft-safe published forms with explicit publish and discard
Stage design edits for already-published forms as drafts so live forms do not silently drift while someone is still editing.
Read more+
Permission-group templates and endpoint labeling for subscribers
Publish standard permission groups, bulk-edit endpoint rights, and expose human-readable endpoint labels instead of raw backend names.
Read more+
CDC-style downstream change pipelines and telemetry reports
Feed external services, analytics, sync workers, and rollout health systems with ordered change events and client-side runtime reporting.
Read more+
Cache-aware public APIs with sanitized rich text and stricter error contracts
Expose selected endpoints publicly with explicit cache rules, sanitize only what is safe to trust, and return cleaner error structures for integrators.
Read more+
