Skip to main content

How the rules are enforced

Every platform makes promises about what its agents will and won't do. The question worth asking is whether a promise is enforced or merely followed - because a rule that depends on every part of a product remembering it is a rule that survives exactly until something forgets.

Yekar.AI holds each guarantee at the lowest layer that can hold it, and a lower layer never defers to the one above it.

On the record itself​

The strongest guarantees are rules on stored records, enforced on every write no matter which part of Yekar.AI made it:

  • Agent memories are append-only. A memory can be superseded by a newer one; it cannot be rewritten to say something it never said.
  • Published Setup revisions are immutable. The version an agent ran last Tuesday is the version you can read today.
  • A published agent or flow cannot be un-published. You can repoint it at a different revision; you cannot make live work vanish.
  • Ownership can't drift across organizations. Every record carries its workspace, and a record can only reference records of the same workspace, so nothing can end up owned by someone else's.

These protections apply to every change made through Yekar.AI, whichever surface made it.

At the moment of action​

Permissions are checked when work starts and when an action is performed. Queued work retains the identity and authorization it was accepted with, and replaying an execution preserves its recorded decisions.

In the message you get back​

Before the hard stop beneath it, a refusal tells you something specific and actionable - which credential is missing, which role you need, which state transition isn't allowed. The hard stop is the guarantee; the message is what turns "no" into "here's what to fix".

Why layer at all​

Each layer answers a different question. A rule on the record proves a shape can never be stored; a permission check proves who was allowed to act just now; a named refusal tells you what to change. A change has to satisfy all of them that apply - which is what lets Yekar.AI say a guarantee holds rather than that it usually does.

Where to go next​