pks status

pks status is a single leaf command registered directly under the root pks command — it renders a system-status dashboard in the terminal using Spectre.Console. It has no subcommands.

Overview

pks status prints a colored per-environment health grid (PROD, STAGING, DEV) followed by a fixed five-row services table (API Gateway, Database, Cache Redis, AI Service, Analytics). With --watch it switches to a live-refreshing panel of CPU, memory, network, request, error, and latency numbers.

Note. Every value this command prints is hardcoded or randomly generated inside the command class. It does not call any real API, read any config file, or check any actual pks-managed resource — no Coolify deployment, no runner, no registry. Treat it as a rendering demo, not a diagnostic tool.

• Health grid: always reports 12/12 running and healthy for every environment, regardless of what actually exists.
• Services table: five fixed rows with fabricated uptime, version, and health-percentage figures.
• Watch mode: fabricates CPU/memory/network/request/error/latency numbers with System.Random every two seconds.
• AI insights: static prose bullets, not derived from any telemetry.

When to use it

Do not use pks status to check the real state of any pks-managed system — it will not tell you anything true. Its only legitimate uses are previewing the Spectre.Console dashboard layout (grid, table, and live-panel rendering) or sanity-checking that the CLI's rendering pipeline works.

For real status of an actual subsystem, use the subsystem-specific command instead:

• pks github status — real GitHub authentication and repository status.
• pks coolify status — real Coolify deployment status.
• pks registry status — real container/NuGet registry status.
• pks vm status — real VM state.
• pks authenticator status — real two-factor enrollment status.

Prerequisites

None. pks status makes no network calls and reads no credential or config store — it runs identically with no prior setup.

Synopsis

pks status [options]

Options

Examples

pks status

Prints the health grid for all three fake environments plus the fixed services table.

pks status --environment prod

Restricts the health grid to the PROD row. The services table below it is unaffected and still prints all five rows.

pks status --ai-insights

Appends the static AI insights panel to the default dashboard.

pks status --watch

Switches to a live panel that refreshes fabricated CPU, memory, network, request, error, and latency numbers every two seconds. There is no exit or iteration-count flag — stop it with Ctrl+C.

Troubleshooting

--environment PROD renders an empty grid

--environment is matched with exact lowercase equality against dev, staging, prod, and all. The header line uppercases whatever string you passed for display, but the grid-row lookup does not — passing PROD (or any other casing) prints a PROD header over an empty grid, with no error message. Pass the value in lowercase.

--watch never stops on its own

Watch mode runs an unconditional loop with a two-second sleep between ticks and no built-in exit condition. There is no --count or --duration option. Stop it with Ctrl+C.

The numbers don't match reality

They are not supposed to. Every figure on this dashboard — health percentages, uptimes, versions, and the watch-mode metrics — is hardcoded or randomly generated in StatusCommand. Use one of the subsystem-specific status commands listed in When to use it for a real check.

See also

• pks — command families and the full 57-command surface.
• pks github runner — real GitHub Actions runner registration and status.
• pks agentics runner — real Agentics runner status.