Product
Ask about your fleet in plain words
The Ask page is a chat with the agents of your workspace. The agent reads your fleet and answers in plain words. It runs no command by itself: a change opens an incident that a person must approve.
Free for up to 5 machines. Every machine starts in audit mode: Kosken reads and reports, and changes nothing.
What the agent can read
- Machines
The list of machines, the facts of one machine, its workloads and its metrics.
- Incidents and signals
The incidents, the timeline of one incident, the signals and the monitors.
- Recipes and checks
Which recipes fit a machine, what one does, and read-only checks that run at once.
- Approvals, audit and memory
The requests that wait for a person, who did what, and the past incidents with the fixes that worked.
What the agent can ask for
Only for a person with write access, and only through the same gates as an incident.
- A change
The agent can ask for a recipe to run on a machine. The request opens an incident. Nothing runs before a person approves.
- A task or a maintenance
The agent can create a task that runs on a schedule, or plan a maintenance window.
Your models, one for each agent
You bring your own key for OpenAI, Anthropic, or DeepSeek. Each stage of an incident and each agent of the chat can use a different model, with its own reasoning effort and token limits. An agent with no model of its own uses the model of its stage. Kosken records the tokens of each call, by stage and by incident.
More of Kosken
- Incident responseFrom a failed signal to a root cause report and a checked fix.
- MonitoringMonitors for services, ports, URLs, containers, disks and logs.
- Recipes and approvalsSigned fixes, five gates in code, and a person who decides.
- Tasks and maintenanceScheduled checks, planned changes, and the jobs of each machine.
- Fleet and modesMachines, workloads, and the audit, shadow and live modes.