Features
Every tool you need. Nothing you don't.
StackBlaze gives you a complete platform, deploy, network, scale, observe, and collaborate, without stitching together a dozen different services.
storefront / prod
Phase canvas · 2 changes staged
Deploy
Build & Deploy
Push to GitHub. We build and ship it.
Connect a repo and StackBlaze builds the image, rolls it out, and gives the app a URL. Preview environments are on Pro and above.
- See apps and add-ons on a live canvas
- Build settings detected from your repo
- Edit environment variables in the dashboard
stack‑app / prod
Phase canvas · production
ingress
ingress
*.stackblaze.app
Running
api
web service
storefront.stackblaze.app
Running
postgresql
add-on · Postgres 16
postgresql:5432
Running
redis
add-on · Valkey
redis:6379
Running
Network
Public HTTPS. Private cluster DNS.
Each web app gets a *.stackblaze.app URL and TLS. Apps in the same phase reach each other on the cluster network.
- Private hostnames between apps in a phase
- HTTP, TCP, gRPC, and WebSockets
- TLS certificates on every public endpoint
Scale
Replicas, CPU, and RAM from the dashboard.
Scale an app horizontally or vertically. Autoscaling uses CPU (and optional spend caps on Pro and above).
- Change CPU, RAM, or replica count without a rebuild
- Default tenant region is us-east-1
- Autoscaling between a min and max replica range
Scales on CPU and memory, not queue depth.
Monitor
Logs, metrics, and alerts in one place.
Monitor resource usage, set custom alerts, and track logs. Full visibility from the moment you deploy.
- CPU, memory, response time, and throughput per app
- Slack, Discord, or webhook notifications when conditions are met
- Live log streaming with search, plus log drains
Collaborate
Phases, PR previews, and one-click rollback.
A project has phases (environments). On Pro and above, a pull request can get its own preview URL. Roll back to a previous revision from the dashboard.
- Phases for production, staging, and the rest
- PR preview URLs on Pro and Business
- Rollback to a previous successful deploy
Stack MCP
Stack MCP for your infrastructure.
Connect Stack MCP to Cursor, Claude, or any MCP-compatible client. Diagnose crashes, query logs, and provision services through natural language, no GPUs, just your infra wired for AI.
- Diagnose issues instantly from any MCP client
- Provision services and wire your stack with natural language
- Query logs, metrics, and config without PromQL or grep
“Why is checkout crashing?”
Functions
Functions scale to zero. They are not apps.
A function is a Knative workload. Cron jobs and background workers cover schedules and queues.
- Knative functions, created separately from apps
- Cron jobs for scheduled work
- Background workers for queues and long-running jobs
| Function | URL | Status |
|---|---|---|
| resize-image | fn-resize-image.stackblaze.app | Running |
| send-receipts | fn-send-receipts.stackblaze.app | Running |
| invoice-batch | fn-invoice-batch.stackblaze.app | Idle · scaled to zero |
Full catalog
Everything in the platform
The complete feature set at a glance.
Build & Deploy
- Git deploys
- Framework detection
- Preview environments
- Rollbacks
- Docker
- Build cache
Networking
- Custom domains
- Automatic TLS
- Private networking
- Load balancing
- WebSockets
- Health checks
Scaling
- Horizontal scaling
- Vertical scaling
- Autoscaling
- Sleep mode
- Zero-downtime deploys
- Managed add-ons
Monitoring
- Metrics
- Logs
- Alerts
- Coroot observability
- Log drains
- Kubernetes events
Collaboration
- Team and RBAC
- Environment variables
- Audit log
- API and MCP
- GitHub
- Templates