What runs on its own at your company.
Six automations are on the register. The three that spend your AI keys run from delivery, and you can stop any of them at any time. The other three are not a setting: two run permanently because they spend nothing, the third is paused.
In the console, on every deployment: Settings, then “What runs on its own”. Switching on and off is reserved for owners and administrators; everyone else sees the state without being able to change it.
On the command line: kastel automations list, kastel automations on, kastel automations off.
This screen is conditioned by no plan, and never will be: we do not sell the right to stop a spend.
Nothing is hidden from you, and you keep the hand
Three functions spend your AI keys, and they run from delivery: the weekly contradiction probe, the check on every sync, lesson compilation. The two that spend nothing run too; the sixth, Context Score measurement, is paused. This page names them all, says what each one costs, and lets you stop any of them.
The reason is sovereignty, not thrift, and it did not change meaning when it changed shape. The question that matters - “what does your tool do with our documents when nobody is watching?” - calls for a verifiable answer, not an absence: what runs is listed here, what it spends is metered, and stopping it is one click away. Every toggle is sealed in your audit log, with its author and its date, in both directions.
The flip side is stated just as plainly: these three functions re-read your content on their own initiative, on your keys, and that is what lets your Kastel learn and watch itself. If one of them bothers you, you switch it off here and it does not come back on its own. The one whose absence weighs most is lesson compilation: that is the one that makes it improve.
The six
| Function | What it does | What it spends |
|---|---|---|
| Weekly contradiction probe | Every week, re-reads pages that might contradict each other and flags the real conflicts before a colleague runs into one | your AI keys - runs by default, stoppable here |
| Check on every sync | As soon as a connector brings in new content, checks straight away that it contradicts nothing existing, instead of waiting for the weekly pass | your AI keys - runs by default, stoppable here |
| Statement attribution | Shows, inside answers, who said what and with what certainty | nothing, ever - runs permanently, off the settings screen |
| Deterministic contradiction scan | A mechanical check that spots obvious conflicts of value, type or date. Works even with no AI key configured | nothing, ever - runs permanently, off the settings screen |
| Lesson compilation | Distils what your AIs have learned into reusable lessons. The function whose absence degrades your Kastel most | your AI keys, outside the monthly cap - runs by default, stoppable here |
| Weekly Context Score measurement | Re-measures the Context Score every week so the figure on screen stays fresh | paused: not running today |
This list is closed at six. Three are yours to set and run by default; statement attribution and the deterministic scan run permanently because they spend nothing, and are not on this screen; Context Score measurement is paused. A seventh function that ran on its own and spent your keys would be a decision of its own, not a convenience added in passing - and it would land on this page, never anywhere else.
Switching on, off, and when it takes effect
Every toggle is recorded in your audit log, with its author, and sealed. Changing a setting is a governance decision: it is traced as one.
The effect is not instant everywhere, and your Kastel says so on each line rather than letting you believe otherwise: most switchable functions take effect on the next upkeep pass, within the hour at the latest; the check on sync takes effect at the next connector sync.
Switching off and back on restores the standard cadence, not necessarily the exact value you might have set by hand on a self-hosted Kastel. That value is in the audit line, so it is recoverable.
Two traps, stated plainly
- The check on sync stands alone. It depends on no other setting: you can therefore have the weekly probe off and the targeted check on. That is a legitimate state, not a bug.
- The scan and attribution are no longer settings. These two used to sit behind a toggle; they now run permanently because they spend nothing, and have left this screen. You will not see them here any more: they just run.
What the checks found
The same screen shows what these functions spent and what they returned: how many questions hit a contradiction on the last pass, how many conflicts went into “To approve”, what share of comparisons was re-read from cache and therefore not paid for twice.
A rate is never shown on its own. It always comes with its 95% confidence interval: “X of N questions hit a contradiction, between A and B% in reality”. On a small sample the interval is wide and the screen says so - it narrows pass after pass. A figure without its interval is not a figure you can act on, and your Kastel refuses to present it as one.
You can also run a check on demand, without waiting for the weekly pass. The estimated cost is shown first, as a range, with what is left of your monthly cap; if the upper bound does not fit, nothing starts. The mechanical check runs without a quote: it costs nothing, ever.
What these functions produce - contradictions to settle, attributed statements, the Context Score - is read and worked on from your Kastel's day-to-day upkeep.