Write the rule once. It runs on every client.
The work your team repeats by hand for every client becomes a rule: when this happens, only if that holds, then do this. It starts by only reporting what it would have done, and it acts once you have read the record and said so.
- 59
- triggers a rule can wait on
- 33
- starter rules, ready to install
- 4
- action tiers, by cost of a mistake
- A risk is created
- CVSS score is at least 9.0
- Detection confidence is confirmed
- The asset is internet-facing
- Open an investigation
- Notify the client's admins
Rules raise the risk. Automations act on it.
Both read the same conditions and return the same evidence. What separates them is what happens when the condition holds.
Seven subjects, twenty-one operators, nested AND and OR groups.
Anything the platform can see, a rule can test.
Runs on a schedule over inventory and posture. When the conditions hold, a risk is raised with the evidence that raised it. That is all a rule does, on purpose.
A risk, with its evidenceTrigger, condition, action.
Waits on an event or a time you set, reads the entity fresh at that moment, and can then notify, ticket, patch, isolate or run an approved script.
An action, on the record- Device
- Microsoft 365 user
- Microsoft 365 application
- Microsoft 365 finding
- Certificate
- Microsoft 365 tenant
- Firewall
Three-valued logic throughout: missing data is not a passing check, and a rule abstains where it cannot see.
Every action, sorted by what it costs when the rule is wrong.
Picking an action from a dropdown is not consent to run it. Anything past a notification needs an explicit opt-in on the rule, and the harder an action is to undo, the more it needs.
Every action in every tier- TIER 0
Notify, create a ticket, assign, tag, add a note
No side effects outside the console. Nothing to gate beyond leaving report-only.
- TIER 1
Open an investigation, approve or schedule a patch, release an isolation
Reversible. Needs an opt-in recorded on the rule itself.
- TIER 2
Isolate a device, run an approved script, revoke sessions, disable an account
Disruptive. Needs the opt-in and fourteen days of report-only history, or a person approving every run.
- TIER 3
Propose accepting a risk, close a risk, snooze a risk
Hard to reverse. Needs the opt-in and a named owner who is a real, active user.

Built on the assumption that a rule will eventually be wrong.
Every guard exists to bound what a wrong rule costs, and all of them are on by default.
- REPORT-ONLY
Every new rule starts here. It evaluates and records, and changes nothing.
- APPROVE
A match holds its actions until a person runs them with one tap.
- ENFORCE
Acts on its own. Getting here is a separate, audited decision.
- 25
- entities in one run
- 20
- runs per hour
- 25%
- of the fleet, never more
- 48h
- older events are ignored
A kill switch, global and per client
Stop every automation across the platform, or for one client, without editing or deleting a rule.
Nobody approves their own rule
The author of a rule cannot approve its runs. There is no admin exception.
Every evaluation is written down
Matched or not, with the values read and any guard that stopped it. A rule that quietly stopped working does not look like one with nothing to do.

Your own scripts, run three ways.
From a weekly cleanup to a fix for a service that keeps stopping, your team's scripts run from one library across every client, and only after a second administrator has approved them.
By hand
Run script in the device menu. Pick an approved script, read its full text, confirm it runs as SYSTEM or root, and see its output.
Scripts switched on for the client, device active and on the right platform
On a schedule
Daily, on chosen weekdays or on a day of the month, at a time in a named time zone, over a client, a device group, named devices or tags. A platform-wide rule runs over chosen clients.
Exactly once, never late; one approval per run in approve mode
From a rule
The run_script action runs a script on the trigger's device, or the case's device for a case. A chain runs up to four more scripts, each only after the one before exited cleanly.
Tier 2: report-only first, then a per-rule opt-in
- A second administrator approves every script, never its author
- Editing a script clears its approval
- Off per client until you switch it on
- Every run is a signed order that pins the approved text
Thirty-three rules you can install today.
Each installs through the same path as a rule you build by hand, so scoping, audit and report-only apply in the same way.
The things worth asking first.
Will an automation ever act without us knowing?
No. A rule cannot be created in enforcing mode. It starts in report-only, where it evaluates and records but changes nothing, and promoting it is a separate audited act with a named person behind it. In approve mode nothing runs until a person taps approve. Once enforcing, every execution is still recorded, including the ones that matched nothing, so you can tell the difference between a rule with nothing to do and a rule that has broken.
Can Secuno run our own scripts?
Yes. Your scripts live in one library, and each runs only after a second administrator, never its author, has approved the exact text. Remote scripts are off per client until you switch them on, and a device owner can refuse them locally. Run one by hand from the device menu, on a schedule across a client or a group, or from an automation rule, which starts in report-only like every other disruptive action. Each run is a signed order pinned to the script's SHA-256 that expires after two hours, and the output is kept for 30 days.
How does this work across a portfolio of clients?
Every rule states its scope explicitly, either one client or the whole portfolio, so a rule written once covers every estate you manage without being copied per tenant. The kill switch works at both levels too: you can stop automation for a single client who is mid-migration without touching anything else.
What happens when a rule turns out to be wrong?
The ceilings decide how much it costs. At most twenty-five entities in one run, twenty runs an hour, and never more than a quarter of the fleet, so the damage is bounded before anyone has noticed. Flip the kill switch and everything stops without deleting a rule. The execution log then tells you exactly which entities were touched, which predicate matched, and what value it read at the time.
The routine work ran itself. The record says how.
Bring us a rule you keep applying by hand. We will build it with you, run it report-only against something real and read the log together.