02 · See · Problems nobody reported

Platform Advisor

Find what quietly went wrong, and give each problem to someone who can fix it.

Most governance problems are not events. An owner leaves, a policy points at a department that no longer exists, a system stops syncing. The Advisor runs checks over your whole estate on a schedule and routes every finding to a named person. It almost never changes anything by itself.

What changes

What you get

01

Checks that run on their own schedule

02

Every finding goes to someone who can act

03

A finding that returns reopens the same record

04

A fix runs only when an administrator asks

Screenshot of the product · Fictional demo data
How it works

It works in three steps

Each step records its reason, and the next step uses it.

  1. Scan

    Each check reads the estate and changes nothing.

  2. Route

    A finding goes to the owner, the security team or the IAM team, and says so when it falls back.

  3. Decide

    A person claims the finding and fixes it where the object lives, or snoozes it until a date.

What it checks

More than 80 checks in five groups

Each check reads your estate on its own schedule and changes nothing. You switch a check on or off and set its threshold.

From the first week

  • A quick scan finds agents without an owner, role candidates and missing contexts
  • Gaps that every new estate has go to setup, not to the inbox
Security and risk · 24Segregation-of-duties conflicts, concentrated privilege, active accounts of people who left, keys that are unused or overdue for rotation, AI agents with too wide a scope.
Configuration · 25Objects and accounts without an owner, owners who left, tasks waiting on someone who cannot log in, rules and policies that point at something that no longer exists.
Data quality · 15Incomplete HR data, identities without a source, duplicate accounts, accounts that cannot be tied to one person.
Provisioning · 11Systems that stopped syncing, changes that keep failing, an agent that went silent.
Tighter access · 10Accounts that have not signed in for a long time, where the system reports it, people who hold more than their peers, groups with the same access that could become a role.
Who gets it

Every finding lands with the person who can act

01

Security team

Segregation-of-duties conflicts, concentrated privilege, a key whose holder left.

02

Object owner

Access on an account that has not signed in, where the system reports it, an agent key or scope the owner should judge.

03

System owner

Accounts on that system that do not add up, AI agents sharing one account.

04

HR data steward

Missing or contradictory HR data.

05

Manager

Someone who holds more access than the colleagues in the same team.

06

IAM team

Rules, tasks and configuration, and whatever has no owner to find.

You can send a finding to someone else. When no owner is found, the finding says so.

What you do with it

Claim it, fix it, or say why not

Three checks can repair themselves. That stays off until an administrator turns it on.

1

Claim

Whoever claims a finding can act on it. Until then, the fix stays locked.

2

One-click fix, where it fits

About twenty kinds of finding have a fix the platform carries out, always at an administrator’s request. For the rest a person decides, and the finding shows where.

3

Snooze or dismiss

Snooze until a date, or dismiss with a reason. A segregation-of-duties conflict cannot be dismissed: it ends with a revoke or an exception.

4

Ask your assistant

Over MCP an AI assistant reads the findings, for example “which agents have no owner?”. It cannot change them.

What we stand behind

A finding that stops being true stays readable, so an auditor can see when it appeared and when it ended.

Included

Included in every price

There is no module to buy for this. The price depends only on how many identities you govern.

How the price works See it in the product tour

Read by role
Platform Advisor

Find what quietly went wrong, and give each problem to someone who can fix it.