ITOM AIOps | September 2026 Store Release Highlights 🛒

YuvalMamka · 3 hours ago

Hey team! 👋

September is a big one. The headline is: the ITOM MCP Server is now generally available in Claude, ChatGPT, and other MCP clients, with full RBAC and audit. Alongside that, Otto gets its first true alert-handling agent, AI decisions become fully explainable and supervisable, and supervision consolidates into a single surface. Here's everything worth knowing.

ITOM MCP Server Console — ITOM Inside Your AI Tools 🧠

The big one. Operations teams can now reach ITOM's native capabilities — alert data, AI models, CMDB relationships, automation logic, service reliability data, and LEAP-driven automation insights — through natural language in the AI tools they already use.

No more fragile Table API workarounds. A single enterprise-grade channel replaces custom scripting against raw APIs.
Built to survive a security review. Every skill invocation is authenticated, scoped to the user's platform roles, and fully auditable, under ServiceNow's data residency policies — so AI-assisted ops scales without new compliance exposure.
LEAP integration puts automation opportunities and configuration intelligence next to traditional operational context.

Required app: ITOM MCP Server Console (v1.1.3)

ServiceNow Otto for ITOM — Agents That Actually Close the Loop 🤖

Otto moves from assistant to operator.

Alert Verification AI Agent — automates alert closure by checking related incidents and knowledge articles, so operators stop re-triaging alerts that are already understood.
Integration Management Agent — conversational setup and lifecycle management for top pull connectors, including credential creation inside the conversation. Onboarding a monitoring source becomes a chat, not a project.
Advanced KB search and summarization during alert investigation — relevant knowledge surfaces without the manual hunt.
Auto-closure reasoning and expanded explanations for alerts, with updated reasoning headers across all closure scenarios.
Offglide execution enabled for compatible alert skills.

Required app: ServiceNow Otto for ITOM (v2.9.4)

Otto-Powered Connector Onboarding 🔌

Otto now reaches the classic on-prem monitoring estate.

New Otto support for SolarWinds, SCOM, Zabbix, Nagios, and vCenter pull connectors — the long tail of monitoring tools gets the same conversational setup experience.
Install integrations with Otto directly from the Integration Launchpad — no context switch between discovery and setup.
Activate with AI is now enabled for all Health Log Analytics data inputs, extending AI-assisted activation across the full log ingest catalog.

Required apps: Service Operations Workspace Integrations Launchpad (v27.3.5), ServiceNow Otto for ITOM (v2.9.4)

AI Transparency & Agentic Supervision 🔍

If AI is going to close and act on alerts, operators need to see why — and be able to step in.

Closure reasoning with supporting evidence for AI auto-closed alerts, viewable from the Express List preview pane or the alert record, with a one-click reopen. Trust becomes verifiable instead of assumed.
AI Processing Experience for agentic workflows — monitor, supervise, and intervene in agentic workflows from both the Express List Alert Panel and the Alert Record, with step-level visibility into each action and access to the underlying reasoning.
Group-agnostic AI Specialists — AI Specialists can now be configured to handle alerts from all groups, not just specific assignment groups or unassigned alerts. Far less configuration overhead for teams running centralized AI coverage.

Required apps: AIOps Experience / Service Operations Workspace ITOM Apps (v27.2.24)

Supervision Consolidated into Express List 🎯

One surface for supervision instead of two.

The separate "Your Work" / "AI Supervision" tab selector has been removed from the AIOps Supervisor Homepage. The homepage now consolidates into a single main panel, with AI Supervision metrics linking out to Express List.
Direct links from AI Supervision metrics into filtered Express List views — alert-category metrics navigate straight to the matching working set, so supervision starts where the work actually happens.
Direct access to AI Specialist activity and auto-resolved alerts in Express List.

Required apps: Service Operations Workspace Supervisor Dashboard (v1.3.7), AIOps Experience / SOW ITOM Apps (v27.2.24)

AIOps Manager Homepage & Navigation 📊

Better fit for managers who own more than one team.

Multi-assignment-group filtering on the homepage, including tracking for alerts assigned to specific groups or left unassigned.
Charts now aggregate alert and incident data across all selected assignment groups, replacing the single all-groups view — so a manager's homepage reflects their actual span of control.
Persistent navigation to AIOps homepages — Manager and Supervisor homepage links are now always available from the "All" menu, regardless of homepage customization.

Required apps: AIOps Experience / Service Operations Workspace ITOM Apps (v27.2.24)

Connector Setup & Troubleshooting ⚙️

Less guesswork when an integration misbehaves.

Improved Push Connector credential setup — select or generate Basic, API key, or OAuth credentials directly in the setup form.
Last error message, payload, and timestamp surfaced in the UI — troubleshooting a failing push integration no longer means digging through logs.
Pull Connector Test & Save status modal replaces the old loader and toast messages, and the integration can be activated straight from it.

Required apps: AIOps Experience / SOW ITOM Apps (v27.2.24), SOW Integrations Launchpad (v27.3.5)

Health Log Analytics 📈

Broader infrastructure coverage, programmatic onboarding, and quieter alerts.

Tiered Host Label correlation across CMDB CI classes — Host Labels on Kubernetes services and other modern infrastructure now correlate to the right CI via an ordered search across hardware, Virtual Machines, Kubernetes, and API-based fallbacks. Alert enrichment and topology context extend to infrastructure that previously fell through the gaps.
OpenTelemetry ingest sources manageable via ServiceNow-native APIs — register, list, status-check, update, and remove OTel sources programmatically, with MCP tool definitions for agent-driven and Otto workflows. No manual UI steps required for onboarding at scale.
Expiring token email notifications for MID-less data integrations — teams get warned before an integration goes silent.
Certificate policy check option for Splunk Poll data inputs (usemidcertificatepolicycheck), giving admins control over MID certificate policy enforcement.
Redundant keyword detections suppressed in anomaly alerts — when multiple keyword detections hit the same component at the same time, only the first surfaces; message detections take priority over keyword detections for the same component and pattern. Silent, real-time, and a direct reduction in duplicate noise.
Manual test samples are now transient in Source Type Structure, clearing on refresh to prevent contamination of downstream flows.

Required app: Health Log Analytics (v40.5.0)

Questions? Comment here — happy to dig in. 💪