Dashboard

Real-time agent workflow monitoring
Checking... Logged in as
End-to-End Agent Workflow

Watch how ITSM ingests infrastructure logs, analyzes impact, creates a ticket, runs the L1→L2→L3 agent graph, matches runbooks, and tracks SLA — all in one flow.

Workflow Pipeline Ready
Agent Graph — Sub-agents & Actions Waiting for demo…

Run the demo to see Master → L1/L2/L3 → Escalation → Validator with live sub-agent actions.

Execution Steps
Step Output

Click Run Full Demo to execute all 7 steps with the production DB pool exhaustion scenario, or click any step on the left to see its output.

Workflow Complete
Live incident queue
Start the log stream to ingest alerts and fill the process queue below.
idle View All Agents
Executive Summary
Loading incident intelligence…
Process Logs
0
# Status Incident Level Type Service Agent Log Line Updated Actions
No process logs for this filter. Start the log stream to queue alerts for processing.
Loading process log…
Process log not found.

Process Details

Overview
Execution & AI Cost
Loading execution metrics…
Escalation Agent — Human Approval

Remediation is blocked until an operator reviews this escalation.

AI Resolution Suggestions
Loading resolution guidance…
Escalation Decision Recorded
Agent Quality Feedback

Rate this agent's work. Corrections are saved as training examples when auto-import is enabled.

Operator Coaching
Open a completed process log to see coaching hints.
Uncovered Agent Capability
Jira Ticket Open in Jira
No Jira issue linked yet.
Master Agent
Sub-Agents (SR01–SR15)
Raw Log
Processing Pipeline
Runbook Followed
Loading runbook details…
Workflow Outcome
Classification & Process Details
Submit Ticket
All Tickets 0
IDTitleLevelSLAJiraConf.Status
Pending Health Incidents (SR01)
Function Title Severity Detected Status Actions
No pending health incidents.
Pending Detection Incidents (SR02)
Service Title Severity Detected Status Actions
No pending detection incidents.
Pending Human Approvals (HITL)
Ticket Title Level Confidence Actions
No pending approvals.
ROI Summary
Incident Analytics
Autonomy Tracker
Service Topology — Nodes & Edges
production 0 nodes 0 edges
Simulate Incident

Select a scenario to highlight down and degraded services.

Impact Summary
Down
0
Degraded
0
Blast
0
Legend
Load balancer
Service
Database / Cache
Worker

Healthy
Degraded
Down / Root cause
Arrows: A → B means A depends on B
Service Registry
Service Type Tier Dependencies Status
/aws/unified/production idle
0 lines
Press Start to tail the unified production log feed. Errors and alerts are classified and queued automatically.
$ itsm tail --follow --group unified-production
Waiting for log stream…
Processing Queue 0
Queue empty — start stream to process alerts
Queue Stats
processing
Queued
0
Processing
0
Done
0
Gaps
0
Currently processing
All queue tasks 0
No tasks yet
Project Log Types 0 types
Scenario Impact Preview

Select a scenario and preview which services are down or degraded.

Down
0
Degraded
0
Blast radius
0
Infrastructure Impact Map production
Healthy Degraded Down Root cause
Agent Profiling

A plain-language profile of every digital worker in the platform. Pick an agent to see what it does, the capabilities it runs today, and the specialized SR sub-agents under L1/L2/L3 Master. Use Add Capability on any agent to check whether a new skill can be implemented there.

Guard drills verify safety. Routing feedback learns from tier mismatches. Use Import Past Data on each agent to build few-shot examples from real ticket history.
Loading agents…
Specialized Sub-Agents 0

Select L1, L2, or L3 Master above to view SR sub-agents.

Select an agent above to view its profile.

Observability & Continuous Improvement
WeekCompletedQualityGuard PassResolve %CSAT
Custom compliance drills
Loading…
Performance & Scale
Loading performance metrics…
TaskModel TierReasonEst. Savings
Agent Quality Scorecard
Loading quality metrics…

Master Agent Self-Awareness
Loading self-awareness metrics…
Master Agent Memory Graph

Node/edge network view of working memory, shared state, and the audit trail.

Back to Agent Profiling
Tenant Management

Create and manage client tenants with AWS, Datadog, Grafana Loki, and ServiceNow log sources. Data is stored in PostgreSQL. Switch the active tenant from the sidebar — log stream and agents reload automatically.

Active tenant
local
All Tenants
ID Client Region AWS Account Log Groups Jira Teams Actions
Loading tenants…
Active Config Preview
Select a tenant in the sidebar to preview its effective configuration.
New Tenant

Target tenant role to assume. The tenant config uses only these fields; the API host must already have AWS auth.
Comma-separated log group paths or prefixes

Service account token (Bearer auth)

Optional sysparm_query fragment
Enable one or more sources — Log Stream will poll all configured connectors for this tenant.
Jira Configuration Per-tenant Jira project settings and connectivity
Tenant Jira Site Project Key Email Status Actions
Loading…
Discovered Infrastructure Topology
0 services Open full graph
Service Type Tier Deps
Tenant:
SLA Policy by Severity
Severity Resolution Target At-Risk Threshold Description
Active Ticket SLAs
Ticket Severity SLA Progress Remaining Status Level
SOC 2 Coverage
Tenant Quota
Audit Trail
Time Event Actor Resource Action Checksum