One kernel between every agent and everything it touches
Once an agent can act, two questions follow: who is it acting for, and who said it could? OrchKernel answers both, before every action.
Every action passes one gate
When an agent wants to do something, the kernel checks it in a fixed order. The answer is allowed, needs approval or denied, with the reason.
Emails, refunds, record changes: every side effect is a request to the kernel. The agent never calls your systems on its own.
The order matters. A kill switch stops an action before anything else is considered. Risk is weighed against trust last, after your rules have had their say.
- 01 Kill switches
- Is this agent, tool or team stopped?
- 02 Arguments
- Are the arguments allowed for this tool?
- 03 Delegation
- Does the person it acts for allow this?
- 04 Budgets
- Is there budget left for the agent, team and task?
- 05 Rules
- What do your rules say about this action?
- 06 Risk against trust
- Is the risk within the agent's trust tier?
Agents propose. Rules decide. The kernel applies.
A $240 refund waits for an admin, then runs exactly once. An agent never has more than the person it acts for. Every decision lands in one tamper-evident log.
Frequently Asked Questions
Put an agent on real work, with people in charge
Design partners run one real workflow on OrchKernel and work directly with the engineers building it.