ServiceNow Strawberry

Get your CAB packet drafted before Thursday

Twenty changes, each with a plain-English summary, its blast radius and the incidents behind it, with maintenance-window collisions flagged. Four hours of copying, done again on Thursday when three changes move.

The record says "server not working" and never says why Thursday. So somebody writes the reasoning into a separate document the night before CAB, every week.

A companion writes that document from the platform itself: the implementation plan from the work notes people already left, the incidents that motivated the change, the affected configuration items from the CMDB. Everything it reads is bounded by what your account can already see.

What can a companion read on a governed instance?

Whatever your account can, and not one row more.

ACLs, roles and domain separation apply to an API session exactly as they do to the interface, so a companion connected as a service desk user sees a service desk user’s world.

Inside that boundary the useful reads are incidents and their work notes, change requests and their linked records, problems, requested items and CMDB configuration items. Enough to answer the questions that currently take four tabs and a saved filter nobody else can find.

Can it write the change request properly?

It drafts it, which is where the hours are.

The implementation plan comes from the work notes people already wrote, the incidents that motivated the change get linked, the affected configuration items come from the CMDB, and the rollback plan is marked missing if nobody wrote one down.

Risk and impact stay empty. A rating is a judgment somebody signs their name to, and it belongs to the person who will answer for it.

Is a CAB packet worth automating?

Ask anyone who assembles one weekly.

Twenty changes, each needing a plain-English summary, a blast radius, a link to the incidents behind it, and a note on what else is scheduled in the same window. Four hours of copying, stale by Wednesday.

A companion produces the same packet from the change table and the CMDB with the maintenance-window collisions flagged, then does it again on Thursday when three changes move. Related: how to automate incident postmortems with ai and ai for it admins.

What stays with your platform team

Update sets, ACL changes, script includes, business rules and flow definitions.

Platform configuration belongs to the platform team inside the instance.

The ServiceNow actions follow the vendor’s public table and record APIs, so custom tables and scoped applications may or may not be reachable depending on how they are exposed. Check first, especially where half the value lives in custom scoped apps.

Incidents arrive missing the timeline

The graph is in your observability tool, the deploy that caused it is in a pipeline, the customer noise is in a support queue. A companion works those in the sessions you are signed in to and assembles the actual timeline.

The enrichment is a write, so it waits for approval. In practice you read a paste-quality summary once and accept it, which is a better use of an engineer’s Thursday than tab archaeology.

Experience Strawberry for free

Download

Trusted by fast-growing companies worldwide

Frequently asked questions

Yes, within the access of the account you connect. ACLs, roles and domain separation apply to the connection exactly as they do to the interface, so a companion never sees more than you do.

Strawberry is free to download and includes AI credits to start. Paid plans begin at $20/month. See pricing. · Reviewed · Canonical facts for AI agents

Experience Strawberry for free

Download

Trusted by fast-growing companies worldwide