Covered On-Call Escalations
Ask who's on call right now, when your next on-call shift starts, or which incidents fired overnight — answered in chat. Acknowledge, respond to or resolve an incident — you confirm each one first.
Where On-Call Coordination Breaks Down
On-Call Management Agent attacks the specific failures that turn a clean rotation into a hunt for the right person — without changing the paging or escalation rules the On-Call app already enforces.
"Who's On Call Right Now?" Becomes A 5-Minute Hunt
The customer ticket is hot. Support pings the team channel. Three engineers reply "not me — I think it's Jamie?" Jamie was last week. The actual on-call is logged somewhere in a scheduling tool nobody bookmarked. The customer waits while we figure out who to escalate to.
Engineers Forget They're On Call Until The Page Fires
Their shift started Monday 9 AM. By Wednesday afternoon they've forgotten — they've scheduled a late dinner, started a long meeting, gone to the gym. The page wakes them up. A quick "when am I next on?" lookup would have spared the conflict.
Nobody Knows Who The Backup Is
The primary is in a meeting and the page needs a second pair of hands. Is Rina the backup tonight, or was that last week? And does her shift end at six or at nine? Without a quick answer, the team guesses — or pings everyone.
Monday's Review Lacks Weekend Incident Context
The weekend is over. The review is Monday. Nobody remembers exactly which incidents fired, which are still open, or how many people answered each page. The incident list is in a tool nobody opened over the weekend.
Open Incidents Hide In A Tab Nobody Has Open
An incident was acknowledged Friday night and never resolved. It sits in the Incidents list while the team assumes someone closed it. A quick "anything still open?" in chat would have caught it.
"Am I On Call Next Week?" Lives In A Calendar Nobody Cross-Checks
An engineer plans a Thursday night dinner three weeks out, forgets to check the rotation calendar, and discovers Tuesday afternoon that he's primary that night. Without a fast "what are my upcoming on-call shifts?" lookup, life-and-rotation conflicts get caught at the last minute — and trades scramble through three Slack DMs.
On-Call Management Agent At A Glance
On-Call AI
Who's on call, your next shift, the open incidents — and the page, taken in chat.
Inside On-Call Management Agent — The Actual Capabilities
Every block below maps to a real tool the agent uses against your On-Call records. Four are reads. Three write — acknowledge, respond, resolve — and each shows the person a preview and waits for them to confirm. The agent never pages anyone and never edits a rotation.
"Who's On Call Right Now?" — Live, Across Every Pool
The single most common on-call question, answered in chat without opening another tool. The agent lists every active pool you can see with the people on call right now — whether each is primary or backup, and when their shift ends.
who_is_on_call— every active pool with the people on call right now.- Filter to a specific pool — partial-match on pool name so "sre" returns both Primary and Secondary.
- Primary or backup, and on call until — so you know who to try next and whether the answer changes in an hour.
- Permission-aware — anyone in the business can ask; pools at locations they can't see are left out.
Your Upcoming Shifts — Today, This Week, This Month
Engineers can ask "when am I next on?" or "am I on this weekend?" and get the answer in chat. The tool returns your on-call shifts from the start of today, in your own time zone, so you can plan around them.
get_my_on_call_shifts— the current user's on-call shifts from the start of today.- Time period filter — today, the next 7 days, or the next 30 (defaults to 7).
- Shows pool and role — the pool, the start and end, and whether you're primary or backup.
- Self-scoped — returns the requesting user's shifts. To see who is on right now, use who_is_on_call.
Pools — What Exists And What's Open
Managers and team leads can list every pool they can see, with how many people are on call in each right now, how many rotations it has, and how many of its incidents are still open.
list_on_call_pools— every pool with its location, who's-on-now count, rotation count, and open incidents.- Active-only filter — default
active_only: true, but you can include retired pools when needed. - Up to 50 results — covers every pool a real engineering org maintains.
- Read-only on rotations — rotation and shift changes still happen in the On-Call app.
Activations — Incident History During A Window
Pull the incidents for a window — used for post-shift reviews, the Monday catch-up, and weekly reliability reports. Filter by status (active, acknowledged, resolved, cancelled), severity, and date to focus on what matters.
list_activations— on-call activations (incidents) with optional filters.- Status and severity filters — active, acknowledged, resolved or cancelled; low, medium, high or critical.
- Date filters — raised or updated before or after a date you name.
- Each incident's answers — who raised it, when, and how many paged people answered and said they could respond.
Outcomes Teams Can Measure
The agent is built to compress "who's on call?" answer time, keep people from forgetting their own on-call shifts, and make incident history quick to pull for reviews. Measure against your pre-agent baseline.
- Time to answer "who's on call?" — seconds from question to definitive answer, replacing channel polling.
- Time to acknowledge — minutes from page to acknowledgement when the responder takes it in chat.
- Post-shift review prep time — minutes to pull the incident list for a specific window.
- Incidents left open — acknowledged-but-unresolved incidents caught by an "anything still open?" check.
- Agent adoption — share of "who's on?" / "when am I on?" questions answered through chat vs scheduling-tool open.
Three Confirmed Writes · Paging And Rotations Stay In The On-Call App
The agent can acknowledge an incident, record your response and ETA, or resolve an incident — and each one shows you a preview of the change and waits for you to confirm before anything happens. Resolving additionally requires the On-Call resolve permission, the same check the app's own button enforces. The agent never pages anyone and never edits a rotation; those stay in the On-Call app where the escalation rules live.
- Three writes, each confirmed — acknowledge, respond and resolve show a preview and wait for you; nothing is written on the model's say-so.
- Resolve is permission-gated — only someone who already holds the On-Call resolve capability is offered the tool at all.
- No paging and no rotation changes through chat — the agent never starts a cascade; who covers which shift is changed in the On-Call app.
- Permission-aware — pool and incident visibility rules from the On-Call app are respected on every read.
- Audit trail on every action — every tool call is logged with the requesting user and the tool used; the AI request log keeps detail for 30 days.
WHAT TEAMS TRY INSTEAD
The four alternatives — and why none of them know the rotation, the incidents, and the visibility rules at once
Most operations and SRE leaders reach for one of these four. None of them stick because none of them combine the live rotation, the incident record, and pool-visibility rules in one permission-scoped surface.
ChatGPT or Claude reading a copied schedule
General-purpose AI on a static rotation paste
- Reads the live on-call shifts and open incidents — not a snapshot from Monday's standup
- Honors pool and incident visibility rules — never returns pools or incidents the user shouldn't see
- Can take the page for you, after you confirm — a copied schedule can only be read
PagerDuty AIOps, Opsgenie AI, Splunk On-Call AI, incident.io AI
Vendor-trapped AI inside the paging platform
- Works for the frontline ops, field service, and operations crews who never had a PagerDuty seat
- Answers from the same people directory and schedules the rest of MangoApps uses — not a separate paging roster
- Same chat surface the rest of the team already uses — no separate ops portal, no extra mobile app
A custom rotation script + Slack reminder bot
Engineering's "let's track on-call ourselves" build
- Already shipped — no cron job to maintain, no Slack bot to keep alive through API deprecations
- Pool-visibility rules enforced at the tool layer — not by hoping the script filters correctly
- Every tool call logged with the requesting user and the tool; the AI request log keeps detail for 30 days
The manual fallback — "who's on call?" Slack message
A Slack ping and a wiki page that's three months stale
- Deflects the routine "who's on this weekend" pings that hit the ops lead a dozen times a day
- Returns the right answer the first time — not the stale wiki everyone forgets to update
- Names the backup and when each shift ends — not just the primary's name
PLATFORM ADVANTAGE
On-Call Agent inherits everything the platform already runs
A standalone on-call vendor has to plumb each of these. The agent gets them for free because the platform already does.
Live rotation records
Reads the same on-call shifts and incidents the On-Call Management app runs on — no parallel store, no drift between the bot and the record.
Pool visibility rules
Location and incident visibility rules respected on every read — people never see pools or incidents they shouldn't.
Confirmed writes only
Acknowledge, respond and resolve each show a preview and wait for your confirmation. The agent never pages anyone and never moves a rotation.
Your day, your time zone
"Am I on call today?" is asked about your day in your own time zone — not the server's.
Audit trail
Every read is logged with the requesting user and the tool; the AI request log keeps detail for 30 days.
Automatic model tiering
The platform assigns each agent one model tier suited to its work — set for the agent, not picked per question.
INDUSTRY FIT
Industries where embedded on-call awareness moves the most weight
On-Call Agent matters most where the rotation is the operations backbone and the answer can't be "let me ask Slack."
Healthcare
Answers who the hospitalist, charge nurse, or specialty on-call is right now — and who backs them up — without a call to the switchboard.
Manufacturing
Surfaces plant on-call coverage and maintenance incidents still open, in one ask from the floor.
Field Services
Lets after-hours technicians check their own on-call shifts and take a page from chat on the road.
B2B SaaS & Tech Ops
Tracks SRE and platform-on-call rotation alongside incident history without a separate PagerDuty seat for everyone who needs to know.
Hospitality
Surfaces property-level manager-on-duty rotation and after-hours coverage without a separate ops portal nobody checks.
Public Sector & Utilities
Agent activity is fully audit-logged. AI requests go to the platform's configured model providers.
WHY MANGOAPPS WINS
An embedded on-call agent beats a chatbot, a paging-platform add-on, or a custom build on every axis
The argument ops, SRE, finance, and IT all share — and the one a horizontal AI or single-vendor add-on structurally cannot answer.
Cheaper than the alternatives
No PagerDuty seat for everyone who needs to know, no Opsgenie SKU, no per-seat ChatGPT, no Slack-bot maintenance — and the ops lead stops fielding "who's on call?" pings.
More secure
Confirmation-gated writes, pool-visibility enforcement, and a logged tool call for every request mean rotation data never leaks beyond the people who should see it.
Easier to deploy
Already deployed if On-Call Management is enabled. The agent is on unless an admin turns it off in On-Call settings, and it reads the pools you already configured.
Easier to use
Lives in chat next to the team that needs to know — no separate paging portal, no mobile app install, no wiki page to keep current.
Easier to manage
Per-business pool rules and visibility scoping sit in the same admin console as every other app's settings.
Easier to extend
Shares the agentic tool framework with every other MangoApps agent. New on-call questions ship as tools, not rewrites.
AI is actually better
A horizontal AI can repeat a pasted schedule. On-Call Agent reads who is on call this minute — primary or backup, and until when — and can take the page for you once you confirm.
At our head office within HR, marketing, and finance, we use ChatGPT quite a bit, and MangoApps AI Helper is a useful tool for us. As I go around to the stores, I’ve been showing them how the AI Intranet Helper can find things and answer plain language questions about our resources, and they’re quite excited about that.
Alison van Zijl
Head of Human Resources
Full House
Customer Success
MangoApps Customers
Organizations that trust MangoApps to run every workflow across their workforce.
Customer Case Studies
Customer Case Studies
Customer Case Studies
Customer Case Studies
Customer Case Studies
Customer Case Studies
Frequently Asked Questions About On-Call Management Agent
It reads four things — who's on call right now across every pool you can see (or one pool you name), each person's role and when their shift ends; the pools with how many people are on now, their rotations and open incidents; your own on-call shifts for today, the next 7 days or the next 30; and incidents filtered by status, severity and date. It can also acknowledge an incident, record your response and ETA, and — if you may resolve incidents — resolve one, each after you confirm.
It can acknowledge, and it cannot page. Acknowledging an incident, recording your response and ETA, and resolving an incident are the agent's three write tools, and each one shows you a preview of the change and waits for you to confirm before it happens — resolving also requires the On-Call resolve permission. The agent never starts a paging cascade and never changes a rotation; those stay in the On-Call app.
Not through the agent. get_my_on_call_shifts is self-scoped — it returns the requesting user's own shifts. who_is_on_call shows who is on call right now and until when; for a pool's upcoming shifts, open the pool in the On-Call app.
list_activations accepts status (active, acknowledged, resolved, cancelled) and severity (low, medium, high, critical) filters, dates the incident was raised or updated before or after, plus a limit (default 20, max 50). Useful for post-shift reviews and weekly reliability reports.
Time to answer "who's on call?", time to acknowledge from chat, post-shift review prep time, incidents left open, and agent adoption versus scheduling-tool opens. Compare against your pre-agent baseline.
Let's Talk
Every team. Every employee. Every workflow. One AI platform — built for the frontline. Since 2008, trusted by 2 million+ users with an NPS of 78.
Why Choose Us?
- Frontline AI: Governed AI for every employee and workflow.
- Top Security: HITRUST, ISO & SOC 2 certified.
- Exceptional UX: Delightful on mobile and desktop.
- Proven Results: 98% customer retention rate.
Rolled out to every employee at AutoZone (125,000), PetSmart (50,000+), A.S. Watson and Raley's (20,000) — and at larger retailers we are not permitted to name.
Prefer to explore first? Ask AI about On-Call AI Agent →