Dashboard
Overview of your organization's resources, fleet health, server metrics, recent deployments, and activity feed.
The Dashboard is the main overview screen for your organization. It aggregates counts, health signals, and recent events from all resources in a single view so you can assess the state of your deployment at a glance.
Overview stats
The stats widget counts every resource that belongs to your active organization:
| Field | Description |
|---|---|
serversDeploy | Servers with type deploy |
serversBuild | Servers with type build |
projects | Projects |
applications | Applications (across all environments) |
s3 | S3 / storage destinations |
registry | Docker registries |
environments | Environments |
compose | Docker Compose stacks |
schedules | Scheduled jobs (app, compose, server, and owner-level) |
members | Organization members |
applicationsRunning | Applications with status running |
applicationsStopped | Applications not running and not in error |
applicationsError | Applications with status error |
The plan field is populated only on Skitflow Cloud — it reflects the owner's current billing tier (core, pro, expert, or enterprise). On self-hosted instances it is always null.
API endpoint: GET /api/trpc/dashboard.stats
Fleet health
The fleet health widget queries the live status of every service across applications, Compose stacks, and all database types (PostgreSQL, MySQL, MariaDB, MongoDB, Redis) and maps each to one of four health states:
| Health | Meaning |
|---|---|
healthy | Status is running or done |
down | Status is error |
unknown | Status is empty, idle, building, or a transient running-* state |
degraded | Any other non-running, non-error state |
Counts are computed over the full visible set before capping the list at 300 services. A service past the display cap still contributes to the summary counts in the header.
Every member with the member role only sees services listed in their accessedServices grant (empty unless they have the services.read permission). Counts reflect that filtered view, not the org-wide total. Owners and admins always see the full fleet.
API endpoint: GET /api/trpc/dashboard.fleetHealth
Server metrics
The server metrics widget fetches live time-series data (CPU, memory, disk, network) from each server's monitoring agent. Up to 20 servers are queried in parallel; each fetch times out after 2 500 ms. Each series returns the last 20 data points.
The combined network value is networkIn + networkOut in MB, matching the per-server detail view.
Requires the monitoring agent to be running and reachable. Configure it under Remote Servers → <server> → Setup Server. See Monitoring for setup instructions.
API endpoint: GET /api/trpc/dashboard.serverMetrics
Recent deployments
Shows the last 12 deployments across all applications and Compose stacks in the organization.
Each entry includes the deployment ID, status, application/compose name, project name, and duration in seconds (null when the deployment has not finished).
API endpoint: GET /api/trpc/dashboard.recentDeployments
Recent activity
The activity feed is an Audit Logs Enterprise feature, available in both cloud and self-hosted deployments (self-hosted requires an Enterprise license key), plus the audit_logs.read permission. It returns an empty list for Core, Pro, and Expert tiers.
Shows the last 12 org-scoped audit log events. Platform-admin events (tier changes, granted-server overrides) are never included.
Each event carries:
type— one ofdeployment,server,backup,user,settingsstatus— one ofsuccess,warning,error,info(derived from the audit action)action,resourceName,userName,createdAt
API endpoint: GET /api/trpc/dashboard.recentActivity