Reflex - Kubernetes Auto-Remediation

Choose your plan

Start free with observe mode. Upgrade to unlock the full trust ladder with intelligent AI root cause analysis.

Free

$0USD

Perfect for getting started with SLO monitoring and learning Reflex automation.

Unlimited Kubernetes clusters
3 SLO definitions
3 Reflex automations
Observe mode only
Curated breach pattern recommendations
Slack notifications
SLO auto-discovery CLI
Prometheus metrics (/metrics)
Community support & docs

Get started for free

Pro

$149/month

Unlock the full trust ladder with unlimited automations and intelligent AI-driven incident analysis.

Everything in Free, plus:
Unlimited SLOs & Reflexes
Dry-run mode with Slack approval buttons
Auto mode with full execution
AI-powered root cause analysis (BYOK OpenAI)
Slack interactive approval (Approve / Reject / Snooze)
Event logging for breach analytics
Priority email support

Frequently asked questions

What is the trust ladder?: The trust ladder is Reflex's unique approach to automation safety. Start in observe mode to learn what actions would be taken, graduate to dry-run mode for approval-based execution, then finally auto mode for immediate responses. Build confidence gradually.
How does AI root cause analysis work?: Pro tier includes intelligent analysis of SLO breaches that goes beyond basic alerts. Instead of just "SLO breached", you get contextualized insights like: "Breach caused by deployment spike. Pattern: bad deploy detected. Recommendation: rollback to api-gateway:v2.1.3. Confidence: 94%". Our AI analyzes patterns, workload types, and system context to provide actionable recommendations with confidence scores.
Can I cancel my Pro subscription?: Yes, you can cancel anytime from your subscription portal. Your Pro features will remain active until the end of your billing period, then your installation will automatically revert to Free tier limitations.
Is my data secure?: Reflex runs entirely within your Kubernetes cluster. No metrics or cluster data leaves your environment. Only license validation requires external communication. Your data stays in your control.
How is this different from Robusta?: Robusta requires writing custom Python playbooks for each remediation scenario. Reflex is SLO-native — you define a target and an action, and the trust ladder handles when and how to execute. No playbook code to write. Works in 15 minutes.
What if it makes things worse?: That's exactly why the trust ladder exists. Start in observe mode — Reflex logs what it would do but takes no action. Promote to dry-run for Slack-approved execution. Only move to auto mode when you're confident. Each reflex is promoted independently.
Do I need to already have SLOs defined?: No. Reflex includes an auto-discovery CLI that scans your Prometheus metrics and suggests SLO targets based on your actual service performance. Most teams go from zero SLOs to monitored services in under 5 minutes.
How does BYOK (Bring Your Own Key) AI work?: You provide your own OpenAI API key. AI analysis runs inside your cluster — breach context is sent to OpenAI for analysis, but your Prometheus metrics and cluster data never leave your environment. You control the cost via your own OpenAI spending limits.

Get started now

# Add Reflex chart repository
helm repo add reflex https://charts.reflexslo.io
helm repo update

# Install free tier (observe mode, 3 SLOs)
helm install reflex reflex/reflex-runtime \
  --set prometheus.url="http://prometheus:9090"

# Install with Pro license (after purchase)
helm install reflex reflex/reflex-runtime \
  --set license.key="YOUR_LICENSE_KEY" \
  --set prometheus.url="http://prometheus:9090" \
  --set slack.webhookUrl="https://hooks.slack.com/services/..."

# Auto-discover SLOs from your Prometheus metrics
# (download from GitHub Releases)
reflex setup --prometheus-url http://localhost:9090