The PromptGuard dashboard provides real-time visibility into your AI application’s security events, usage patterns, and performance metrics.
Dashboard Overview
Access your monitoring dashboard at app.promptguard.co to view:- Real-time security events and threat detection
- Usage analytics and API consumption
- Performance metrics and latency monitoring
- Error rates and system health
- Cost tracking and billing information
Security Monitoring
Threat Detection
Monitor security events in real-time:- Blocked Requests: Requests stopped by security policies
- Redacted Content: PII and sensitive data automatically masked
- Attack Patterns: Common injection and jailbreak attempts
- Risk Scores: Threat level assessment for each request
Security Analytics
Track security trends over time:- Threat Volume: Number of security events per day/hour
- Attack Types: Distribution of different threat categories
- False Positives: Requests incorrectly flagged as threats
- Policy Effectiveness: Success rate of security policies
Event Details
For each security event, view:- Request Content: Original prompt (sanitized for display)
- Detection Reason: Why the request was flagged
- Risk Score: Numerical threat assessment
- Response Action: Block, redact, or allow decision
- Timestamp: When the event occurred
Usage Analytics
API Metrics
Track your API usage across all endpoints:- Request Volume: Total requests per time period
- Model Usage: Distribution across different AI models
- Provider Split: Usage between OpenAI, Anthropic, etc.
- Success Rates: Percentage of successful requests
Cost Tracking
Monitor AI costs and usage:- Token Consumption: Input and output token usage
- Cost per Model: Spending breakdown by AI model
- Monthly Trends: Cost patterns over time
- Budget Alerts: Notifications when approaching limits
Geographic Distribution
View usage patterns by location:- Request Origins: Geographic source of API calls
- Latency by Region: Performance across different locations
- Regional Security Events: Threat patterns by geography
Performance Monitoring
Latency Metrics
Track response times and performance:- Average Latency: Mean response time across all requests
- P95/P99 Latency: High-percentile response times
- PromptGuard Overhead: Additional processing time
- Provider Latency: Underlying AI service performance
System Health
Monitor overall system status:- Uptime: Service availability percentage
- Error Rates: Failed request percentage
- Rate Limits: Current usage against limits
- Capacity: System load and performance
Performance Trends
Analyze performance over time:- Daily/Weekly Patterns: Performance trends
- Peak Usage: High-traffic periods
- Degradation Alerts: Performance issue notifications
Alerts and Notifications
Security Alerts
Configure notifications for security events:- High-Risk Requests: Immediate alerts for severe threats
- Unusual Patterns: Notifications for anomalous behavior
- Policy Violations: Alerts when rules are triggered
- Failed Authentication: Invalid API key usage
Performance Alerts
Monitor system performance:- High Latency: Response time threshold alerts
- Error Rate Spikes: Increased failure notifications
- Rate Limit Warnings: Approaching usage limits
- Downtime Alerts: Service availability issues
Usage Alerts
Track consumption and costs:- Budget Thresholds: Spending limit notifications
- Usage Spikes: Unusual activity patterns
- Token Limits: Approaching consumption caps
- Billing Alerts: Payment and subscription issues
Custom Dashboards
Dashboard Widgets
Customize your dashboard with:- Metric Charts: Time-series graphs of key metrics
- Status Indicators: Current system health status
- Top Lists: Most active users, models, endpoints
- Trend Analysis: Period-over-period comparisons
Filtering and Segmentation
Filter dashboard data by:- Time Ranges: Last hour, day, week, month
- API Keys: Specific application or environment
- Models: Filter by AI model type
- Security Policies: View by policy configuration
Reporting
Automated Reports
Schedule regular reports:- Daily Security Summary: Security events and trends
- Weekly Usage Report: API consumption and costs
- Monthly Analytics: Comprehensive usage analysis
- Custom Reports: Tailored to your specific needs
Export Options
Export data for external analysis:- CSV Export: Raw data for spreadsheet analysis
- JSON API: Programmatic access to metrics
- PDF Reports: Formatted reports for stakeholders
- Dashboard Screenshots: Visual snapshots
API Access
Dashboard Analytics API ✅
Access analytics through the dashboard API (requires session authentication):Developer Usage API ✅
Get usage statistics via developer API (requires API key):Roadmap Features 🚧
The following advanced metrics API endpoints are planned but not yet available:Webhook Integration
Receive real-time notifications:Mobile Access
Mobile Dashboard
Access monitoring on mobile devices:- Responsive Design: Optimized for mobile viewing
- Key Metrics: Essential information at a glance
- Push Notifications: Critical alerts on mobile
- Offline Support: View cached data without connection
Next Steps
Usage Tracking
Set up detailed usage analytics and tracking
Alerts & Webhooks
Configure notifications and integrations
Audit Logs
Review detailed activity and security logs
Security Overview
Configure security policies and detection
Dashboard Tips
How often is dashboard data updated?
How often is dashboard data updated?
Most metrics update in real-time, with some aggregated statistics refreshing every 5-15 minutes. Security events appear immediately.
Can I create custom dashboard views?
Can I create custom dashboard views?
Yes, you can customize widget layouts, create filtered views, and save dashboard configurations for different use cases.
How long is dashboard data retained?
How long is dashboard data retained?
Detailed event data is retained for 90 days, with aggregated metrics available for 2 years. Enterprise plans offer extended retention.
Can I integrate with external monitoring tools?
Can I integrate with external monitoring tools?
Yes, use our Metrics API and webhook integrations to send data to tools like Datadog, New Relic, or custom monitoring systems.