Agent

Family & referrer correspondence

A coordinator's inbox is mostly the same administrative questions, each one needing a look at the file before it can be answered. This agent drafts the routine replies using your standard wording and the detail from the connected file, and leaves them in the drafts folder. Anything touching care, a complaint or a safeguarding concern is stopped and handed to a person instead.

How it runs

Step by step

01

Connect the shared mailbox

The coordination mailbox is connected along with the administrative records the replies draw on, and nothing beyond that.

02

Agree the subjects it may draft

Visiting arrangements, document requests, invoicing questions, appointment logistics. Everything else is out of scope by default rather than by exception.

03

Load your standard wording

The agent drafts in the phrasing your organisation already uses, so replies read like your team and not like a template.

04

It drafts the routine replies

Each draft answers the question with the detail from the file, and cites where in the file that detail came from.

05

It stops on sensitive subjects

Care, complaints and safeguarding are never drafted. Those messages are flagged and routed to the named person, unanswered.

06

The coordinator reads and sends

Every reply is read and sent by a person. Nothing leaves the drafts folder on its own, ever.

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

Service coordinators and administrators holding a shared mailbox, in organisations where the same administrative questions arrive every week.

Escalation

When it asks a person

The coordinator reads and sends every reply; nothing leaves the drafts folder on its own

Applications & Rights

Which tools it uses, and what it may do in each
SharePointRead only
Google DriveRead only
GmailReadWrites 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

Service coordinators and administrators holding a shared mailbox, in organisations where the same administrative questions arrive every week.

What you’ll need

  • The shared mailbox connected
  • An agreed list of subjects it may draft
  • Your standard wording supplied
  • A named person for anything sensitive

What you get

  • Draft replies waiting in the mailbox
  • A source reference inside each draft
  • A routed list of sensitive messages

What it doesn’t do

It handles administrative correspondence only. Anything about a person's care, a complaint or a safeguarding matter is stopped and passed to a named person rather than drafted, and it sends nothing itself under any configuration.

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.

Repetitive inbox draftingCopy-paste reply templates
Estimated savingNot setNo figure ships until someone at ollo owns it and the method behind it.

Frequently asked

How is a sensitive message actually detected?

Against the subject list your team writes in the co-build, not a generic classifier. Anything the agent is unsure about is treated as sensitive and routed, because the cost of a wrong stop is a small delay and the cost of a wrong draft is much higher.

Can it ever send automatically?

No. This one is built so that the send is a person's action, and we do not offer an automatic mode for it. If a reply is urgent, it is quicker for the coordinator to press send on a ready draft than to write from scratch.

What personal data does it touch?

Only what you connect, which for this agent is the administrative record rather than the care record. Your information governance lead sets the scope in the build and the rights are per source.

Will families notice a difference?

They should notice replies arriving sooner and reading consistently, because the agent removes the drafting delay rather than the person. Every message they receive has been read and sent by a member of your team.

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.