Approval Queue

AI proposes. Humans approve.
Zero unauthorized changes.

Not every AI action should execute automatically. Nova's Approval Queue puts a human in the loop for high-risk operations, production database changes, security policy modifications, infrastructure scaling beyond thresholds, and anything else you define. The AI does the analysis and recommends the action; your team makes the final call with a single click.

Start Free Trial Watch Demo
app.novaaiops.com · Approval Queue
● LIVE
Nova Approval Queue Dashboard
0
Unauthorized changes
Risk
Risk-scored actions
1-click
Approve / deny
100%
Full audit trail
Risk-Scored Action Cards

Every pending action comes with a risk score and full context

When an AI agent proposes a high-risk action, it doesn't just land in a queue as a cryptic request. Nova generates a rich action card with the agent's reasoning, the affected systems, the risk score (calculated from blast radius, service tier, and time of day), historical success rate for similar actions, and a what-if simulation of the expected outcome. Your approver has everything they need to make an informed decision in seconds.

  • ✓
    Blast radius analysis — see exactly which services, pods, and users will be affected by the proposed action
  • ✓
    Risk scoring model — composite score from service criticality, change scope, time of day, and historical failure rate
  • ✓
    What-if preview — simulated outcome showing expected impact before you approve, so you can catch edge cases
app.novaaiops.com · Action Cards
Risk-scored action cards
Policy-Driven Routing

Define once which actions need approval: and who can approve them

Not every action needs the same level of oversight. Nova lets you define granular approval policies: auto-approve pod restarts for non-production clusters, require senior SRE approval for production database failovers, and mandate VP sign-off for security group changes. Policies route actions to the right approver automatically, with escalation chains for time-sensitive requests.

  • ✓
    Tiered approval policies — different approval requirements by action type, environment, service tier, and risk score
  • ✓
    Role-based approvers — route to specific teams, on-call engineers, or management chains based on action classification
  • ✓
    Auto-escalation — if primary approver doesn't respond within your SLA, automatically escalate to the next person in the chain
app.novaaiops.com · Approval Policies
Policy-driven approval routing
1-Click Approve from Anywhere

Approve from Slack, mobile, email, or the dashboard: one click is all it takes

When a production-critical action needs approval at 3 AM, you shouldn't have to open a laptop and navigate to a dashboard. Nova sends approval requests to Slack, Microsoft Teams, PagerDuty, email, and mobile push notifications. One click to approve, one click to deny, with an optional comment field for context. The action executes immediately upon approval, no additional steps required.

  • ✓
    Multi-channel delivery — approval requests via Slack, Teams, PagerDuty, email, SMS, and mobile app push notifications
  • ✓
    Instant execution — approved actions execute within seconds of approval, with real-time status updates back to the approver
  • ✓
    Denial with feedback — when you deny an action, add a reason that feeds back to the AI agent to improve future recommendations
app.novaaiops.com · Quick Approve
1-click approval from any channel

Goes great with Approval Queue

AI Audit Trail

Every approval and denial is permanently recorded in the audit trail with timestamp, approver, and reasoning.

Agent Ledger

Approval rates and denial reasons feed back into agent trust scores and performance metrics.

Auto Remediation

Low-risk remediations execute automatically while high-risk ones route through the Approval Queue.

AI speed with human judgment: the best of both worlds

See how Nova's Approval Queue ensures zero unauthorized AI actions while keeping response times under minutes.

Start Free Trial Request a Demo