AI Agent Sentinel - Automated Monitoring & Alerting

Marketplace Getting Started Updated March 15, 2026 9 views

AI Agent: Sentinel - Automated Monitoring & Alerting

Sentinel is an AI-powered monitoring and alerting agent that continuously evaluates business rules against real-time workforce data. Define conditions across scheduling, attendance, overtime, leave, and service desk domains, and Sentinel automatically triggers notifications, AI analysis, emails, webhooks, or task creation when thresholds are breached.

What is Sentinel?

Sentinel is the automated monitoring and alerting system inside MangoApps Workforce. It lets administrators define rules that watch key workforce metrics — unfilled shifts, late arrivals, overtime hours, pending leave requests, open service desk tickets — and take action when conditions are met. Actions range from in-app notifications and email alerts to AI-powered analysis that provides severity assessments and recommended next steps.

Sentinel operates through the Workflow Automation Rules engine with a dedicated sentinel trigger type, evaluated on a recurring schedule by a background job. Each rule evaluation is logged as an Autonomous Action for full audit traceability.

Core Value Proposition:

Rule-Based Monitoring — Define conditions with domain, metric, operator, and threshold (e.g., “unfilled shifts next 7 days > 5”)
Multi-Domain Coverage — Monitor scheduling coverage gaps, attendance (late arrivals), overtime hours, leave request backlogs, and service desk ticket queues
AI-Powered Analysis — Optionally trigger LLM analysis that provides severity assessment, root cause analysis, and actionable recommendations
Priority-Based Alerting — Critical, high, medium, and low priority levels with configurable recipient targeting (admins, managers, or specific users)

At a Glance

Rule Engine	Real-Time Metrics	AI Analysis	Notification Actions
Domain + metric + operator + threshold	Scheduling, attendance, overtime, leave, service desk	LLM-generated severity & recommendations	In-app, email, webhook, task creation

Perfect For:

Administrators — Set up proactive monitoring rules that catch workforce issues before they escalate
Managers — Receive targeted alerts about team-level metrics like coverage gaps or overtime spikes
HR Teams — Monitor leave request backlogs and compliance-sensitive metrics automatically

How It Works

┌────────────────────────────────────────────────────────────────┐
│                    SENTINEL MONITORING FLOW                      │
├────────────────────────────────────────────────────────────────┤
│                                                                  │
│  ┌──────────┐    ┌───────────┐    ┌───────────┐               │
│  │  DEFINE   │───▶│ EVALUATE  │───▶│  TRIGGER  │               │
│  │  Rules    │    │  Metrics  │    │  Actions  │               │
│  └──────────┘    └───────────┘    └───────────┘               │
│       │                │                │                      │
│  Admin creates    Background job     Conditions met:           │
│  rule with         gathers live       execute configured       │
│  domain/metric/    metric data        action chain             │
│  threshold         from database                               │
│                                                                  │
│  ┌──────────────────────────────────────────────────────┐     │
│  │              AVAILABLE ACTIONS                         │     │
│  ├──────────────────────────────────────────────────────┤     │
│  │  notify_alert  │  ai_analyze  │  send_email          │     │
│  │  send_webhook  │  create_task │                       │     │
│  └──────────────────────────────────────────────────────┘     │
│                         │                                      │
│                         ▼                                      │
│               ┌──────────────────┐                             │
│               │  AUDIT LOG       │                             │
│               │  AutonomousAction│                             │
│               │  with full       │                             │
│               │  context data    │                             │
│               └──────────────────┘                             │
│                                                                  │
└────────────────────────────────────────────────────────────────┘

Monitored Domains & Metrics

Domain	Metric	What It Measures
Scheduling	`coverage_gap_count`	Unfilled shifts (no user assigned) in the next 7 days
Attendance	`late_count`	Employees who clocked in after their scheduled start time today
Time & Attendance	`overtime_hours`	Total overtime hours across all timesheets for the current week
Leave	`pending_requests`	Leave requests in pending status awaiting approval
Service Desk	`open_tickets`	Service desk tickets in open or in-progress status

Key Features

Rule Configuration

Capability	Description
Domain & Metric Selection	Choose from scheduling, attendance, overtime, leave, or service desk domains
Condition Operators	Greater than, less than, equals, and threshold-based comparisons
Priority Levels	Assign critical (4), high (3), medium (2), or low (1) priority to each rule
Enable/Disable Toggle	Activate or pause individual rules without deleting them
Dry Run Mode	Preview what actions would execute without actually triggering them

Action Types

Action	Description
Notify Alert	Send in-app notifications to admins, managers, or specific users with deduplication (one alert per rule per day)
AI Analyze	Trigger an LLM analysis that provides severity assessment, situation analysis, recommended actions, and urgency evaluation
Send Email	Dispatch email notifications to configured recipients
Send Webhook	POST alert data to external systems for integration
Create Task	Automatically create a task in the Tasks app for follow-up

Audit & Traceability

Capability	Description
Autonomous Action Logging	Every rule evaluation that triggers an action is logged with full context
Idempotency Protection	Duplicate alerts for the same rule on the same day are suppressed automatically
Recent Alerts Dashboard	View the last 7 days of sentinel alerts on the admin index page
Per-Rule Alert History	Drill into a specific rule to see all historical alerts it has generated

User Roles & Permissions

Action	Employee	Manager	HR / Admin	Super Admin
Receive sentinel alert notifications	—	✅	✅	✅
View sentinel alerts dashboard	—	—	✅	✅
Create and edit sentinel rules	—	—	✅	✅
Enable/disable rules	—	—	✅	✅
Run dry-run evaluations	—	—	✅	✅
Delete sentinel rules	—	—	✅	✅

Getting Started

For Administrators

Navigate to Sentinel — Go to Admin → Sentinel Rules to see the rules dashboard and recent alerts
Create a rule — Click New Rule, select a domain and metric, set the threshold and operator, and choose a priority level
Configure actions — Add one or more actions: notify admins, trigger AI analysis, send email, or create a follow-up task
Test with dry run — Use dry-run mode to preview what would trigger without actually sending notifications
Enable the rule — Toggle the rule to active; the background evaluation job will check it on its next scheduled run
Monitor alerts — Review recent alerts on the dashboard and drill into individual rules for history

Best Practices

Start with high-impact, low-noise rules — monitor critical metrics like unfilled shifts and overtime before adding lower-priority alerts
Use AI analysis for complex situations where the metric alone does not tell the full story
Set appropriate priority levels — critical alerts require immediate human attention, while medium/low can be batched for review
Review and tune thresholds regularly as your workforce patterns change with seasons or growth
Use dry-run mode when creating new rules to verify they trigger at the right frequency before going live

Workforce Bots Overview — AI-powered bots for workforce communication and automation
Apps Overview — Browse all available marketplace apps

Sentinel — Proactive AI monitoring that catches workforce issues before they escalate.

Was this article helpful?

Article ID: sentinel_agent_app_overview