Agent

Handbook & policy upkeep

Handbooks rot quietly. Two pages disagree, a process changed and nobody updated the page, and employees stop trusting the whole thing. This agent reviews the policy set against itself and against the questions people actually ask, then drafts the revisions. People Ops approves every wording change before it is published.

How it runs

Step by step

01

Review the whole policy set

The agent reads the connected handbook and process documents, and records when each page was last changed.

02

Find pages that contradict

Pages saying different things about the same rule are listed as pairs, with the conflicting wording shown side by side.

03

Pull the unanswered questions

Questions the policy desk could not answer become the gap list, so the review reflects what employees needed rather than a guess.

04

Draft the proposed revisions

Each change is drafted in your house style, showing the current wording next to the suggested replacement.

05

Prioritise by how often asked

Pages that get asked about most often go to the top of the review list, so the thin ones are fixed first.

06

People Ops approves each change

Nothing is published by the agent. People Ops accept, edit or reject each revision and put the approved wording into the handbook.

The harness

Exactly what this agent can see, touch and change

The same five controls sit behind every Askollo agent. These are this one's settings — visible before you build it, not buried in an admin screen afterwards.

Context

What reaches the model

People leads who inherited a handbook they did not write and cannot tell which parts still hold. Useful before an audit, a policy refresh, or a move into a new country.

Escalation

When it asks a person

People Ops approves each revision before it goes into the handbook

Applications & Rights

Which tools it uses, and what it may do in each
NotionRead only
ConfluenceRead only
SlackRead only
Google DriveWrites newCreates new records or documents. Never edits, moves or deletes anything that was already there.

Verification

How you know it’s right

Every claim links to the document it came from. A statement the agent cannot cite does not make it into the output — which is what makes the result reviewable in minutes rather than re-read end to end.

Who it’s for

People leads who inherited a handbook they did not write and cannot tell which parts still hold. Useful before an audit, a policy refresh, or a move into a new country.

What you’ll need

  • A connected handbook or wiki
  • A log of employee questions
  • An agreed house style
  • People Ops to approve wording

What you get

  • A list of contradicting pages
  • A gap list from real questions
  • Drafted revisions for review

What it doesn’t do

It proposes wording. It does not publish to the handbook, decide policy, or tell you whether a policy is lawful in a given country. That needs your adviser.

How you get it

We build the first one with you

Not a template you configure alone. We sit with your team, build it on real data, and hand over the controls.

01

Scope

One session with the people who actually do the work. We agree what the agent reads, what it may write, and who approves.

02

Co-build

Built on your own data, not a sandbox. You watch it being made, so you know why it behaves the way it does.

03

Handover

You own the controls. Change the context, tighten the rights, move the approval gate — without coming back to us.

What it replaces

Parts of the work currently spread across the categories below. It does not replace any of those products outright.

Handbook reviews done by handPolicy gaps discovered through complaintsConflicting process notes across wikis
Estimated savingNot setNo figure ships until someone at ollo owns it and the method behind it.

Frequently asked

Will it rewrite our tone of voice?

It drafts in the house style you connect and shows the current wording next to the proposal, so every change is visible. Where you have no written style, it keeps the existing sentence shapes rather than imposing new ones.

How does it know a page is out of date?

It uses signals it can actually see: the last change date, a contradiction with another page, and questions employees asked that the page failed to answer. It does not declare a page wrong on its own, it puts the evidence in front of you.

Can it check our policies against employment law?

No. It works from the documents you connect, so it can tell you that two of your pages disagree, but not which one is lawful. Legal review stays with your adviser.

What does the People team still have to do?

Review the drafts and approve the wording. The part that used to be finding the problems is done for you. Deciding the policy is still yours, and we would not build it any other way.

Let's Build

AI is a capability you build. Let's build it together.

30 minutes with our team and you'll leave with a real plan — not a sales pitch.