Inspect a Spinup Agent

Run by agent ID:

spinup agents get agent_01hxyz...

Or run by slug:

spinup agents get support-agent

Control-plane agent commands treat values starting with agent_ as IDs and other values as slugs. Use --slug only when you need to force slug lookup for an unusual value.

Example output:

Workspace: acme
ID: agent_01hxyz...
Name: Support Agent
Slug: support-agent
Deployment: Pending deploy
Active Release: v1 (agent_release_01hxyz...)
Undeployed Changes: 1
Primary Model: openrouter/openai/gpt-4o-mini
Max Output Tokens: 64000
Default Harness: openclaw
Supported Harnesses: openclaw, hermes
Runtime: 4096 MiB memory, 4 GiB disk
Skills: 1 skill
Capabilities: 1 capability
Status: ready
Reconciliation: applied
Reconciliation Error: -
Ready At: 2026-04-17T10:06:00.000Z
Last Error: -
Created At: 2026-04-17T10:00:00.000Z
Updated At: 2026-04-17T10:05:00.000Z

JSON output

Use --json for the full public inspection payload:

spinup agents get support-agent --json

The JSON response includes the agent ID and slug, deployment status, active release metadata, undeployed-change count, model, harnesses, capabilities, runtime sizing and network policy, environment status, environment attention, and runtime reconciliation state. Environment variable values and secret references are not returned.

Configuration commands save draft changes. Runs, schedules, and Agent Chat use activeRelease; run spinup agents deploy <agent> when the current draft should become active.

Choose or override the workspace

Persist the default workspace once:

spinup workspaces use acme

Or override it for one command:

SPINUP_WORKSPACE=acme spinup agents get support-agent