Promote the release. Just ask.Batch, check and promote update sets along your pipeline, in conversation, with your approval before every move.
- Created catalog item New Laptop Request · sc_cat_item · read back
- Reused variable set Laptop Configuration · io_set_item · read back
- Created UI policy “Require accessory details when Other is selected” · catalog_ui_policy · read back
- Created client script “Recommend RAM for laptop model” · catalog_script_client · read back
Build complete — all 4 layers verified. Update set Phyllis - AdHoc - NewLaptopRequest - 2026 left in progress.
- Explored 2 sources · Ran for 33s
Short answer: the P2 “breaches” mostly come from duplicated SLA definitions piling clocks onto a queue nobody owns, not from slow resolution on one clock.
Stacked SLA definitions. Several overlapping SLA sets attach to P2 incidents, so one late P2 shows up as many breaches.
No owner. Of the 36 active P2 incidents, 33 have no assignee.
Wrong SLAs attached. “IT Support P1/P3 — Resolution” clocks are attached to P2 incidents.
- Explored 43 sources, wrote 1 document · Ran for 3m 46s
The assessment is done and the full report is in the Artifacts panel. It scores the CMDB 57/100 (Poor, D).
Nothing on the instance was changed.
- Explored 11 sources · Ran for 3m 40s
A rebuilt, single canonical “Travel Request” catalog item and its full fulfilment lifecycle — order form, tiered approvals, Travel Desk booking and notifications.
33 stories across 7 epics, each with effort, points and acceptance criteria.
- Ran suite SPM — Regression Suite · TES0001108 · 1 minute
The suite ran on acmedev: 8 passed, 0 failed. Each test and its result is open beside the chat, with the run record.
- Previewed on acme-test · No conflicts · health checks clean
Previewed the update set on acme-test — no conflicts. Seven records are captured, all in the right scope.
Promote Phyllis - AdHoc - NewLaptopRequest - 2026 to acme-test? Approve
- Explored 23 sources · Read your team conventions
The Build Book lists the skills and team conventions I’m following, and every reference resolved on your instance before anything is built.
Built for how ServiceNow teams work
- ISO 27001 certified
- Per-user OAuth
- No training on your data
- Scoped update sets
- Jira · Agile 2.0
Batch it, and know it’s ready
Describe and scan each update set, batch them, move stray entries to the right set, and get a ready-for-release verdict before anything moves.
Retrieve, preview, commit
Promote along your pipeline — to test, UAT and staging — with an approval summary first and progress you can follow in the chat.
Preview problems, resolved in chat
Each preview problem is listed and explained. Accept or skip it, then retry the commit — without leaving the conversation.
Releases, in conversation.No new console to learn. Ask, review the summary, approve.
“Batch my three update sets for the onboarding release and check they’re ready.”
Batching“Deploy this update set to UAT.”
Promote“What preview problems came up on test, and what should I do about each?”
Preview“What’s actually in this update set?”
DescribeNothing ships without your approval.
Signs in as you.
Per-user OAuth — she inherits your roles and ACLs, and every change is logged under your name.
ApprovalsApproval on every change.
Approval cards before changes are written — one at a time, or the whole approved build. Nothing reaches production from chat.
ApprovalsPer-instance controls.
Build, Read-only or Backlog only per instance. Table rules and switched-off tools are enforced on every request.
ApprovalsQuestions & Answers
Is there a deployment console?
No — deploying is a conversation. Ask, review the approval summary, approve.
Can Phyllis deploy to production?
Not from chat. She promotes along your pipeline to test, UAT and staging; production stays with your release process.
What happens with preview problems?
Each one is listed and explained. You accept or skip it, and she retries the commit.
Do I need to set anything up?
Once: your pipeline in Phyllis, and an Update Source on each target in ServiceNow. Phyllis uses them — she never creates Update Sources.
Get started with Deploy.
See it on your own instance in thirty minutes.