Dashboard & Trends
Monitor your codebase health with metric cards, trend indicators, and time-series charts
Dashboard Overview
The repository dashboard is your central view for understanding codebase health. It displays metric cards with current values, trend direction, sparklines, and a composite health score. Each card updates automatically after every scan.
Navigate to any connected repository to see its dashboard. If the repository has been scanned at least once, you will see populated metric cards. Otherwise, you will see a prompt to run your first scan.
Key Metrics
The dashboard tracks seven core metrics extracted from each scan.
| Metric | Key | Description |
|---|---|---|
| Avg Complexity | complexity_avg | Mean complexity across all analyzed symbols |
| Max Complexity | complexity_max | Highest complexity value found in the scan |
| P95 Complexity | complexity_p95 | 95th percentile complexity, filtering out extreme outliers |
| File Count | file_count | Total number of source files analyzed |
| Total LOC | total_loc | Total lines of code across all files |
| Symbol Count | symbol_count | Total functions, classes, and methods extracted |
| Alert Count | alert_count | Number of active alerts from the latest scan |
Trend Indicators
Each metric card shows a trend indicator comparing the current value to the previous scan. The trend is calculated as a percentage change.
Increasing (greater than +5%)
The metric has gone up since the last scan. For complexity metrics, this usually indicates a regression that warrants investigation.
Decreasing (less than -5%)
The metric has improved. For complexity metrics, this signals successful refactoring or cleaner new code.
Stable (within +/- 5%)
The metric has not changed significantly. No action needed.
Sparklines
Each metric card includes a 30-day sparkline — a compact mini chart that shows the metric trend over the last 30 days. Sparklines give you an at-a-glance sense of whether a metric is trending up, down, or holding steady, without needing to open the full time-series view.
Note: Sparklines require at least two data points (scans) to render. If you have only scanned once, you will see a flat line until the next scan completes.
Health Score
The health score is a composite 0–100 rating that summarizes your repository's overall code quality. It is calculated from alert counts and complexity metrics.
How It Works
- Starts at 100 (perfect health)
- Critical alerts deduct the most points
- Warning alerts deduct a moderate amount
- Info alerts deduct a small amount
- High average complexity further reduces the score
- Score is clamped to 0 (minimum)
| Score Range | Rating | Interpretation |
|---|---|---|
| 80 – 100 | Healthy | Few or no alerts, low complexity |
| 50 – 79 | Needs Attention | Some warnings or moderate complexity |
| 0 – 49 | At Risk | Multiple critical alerts or very high complexity |
Time-Series Charts
Click any metric card to open a full time-series chart showing the metric's history over time. Charts support multiple time periods for different levels of analysis.
Period Selector
7d
Last 7 days
30d
Last 30 days
90d
Last 90 days
1y
Last year
Data Retention
How long your metric data is stored depends on your plan. Longer retention means more historical context for trend analysis.
| Data Type | Hobby | Pro |
|---|---|---|
| Raw scan data | 30 days | 1 year |
| Weekly aggregates | — | 1 year |
| Monthly aggregates | — | Forever |
Upgrade for more history: Hobby plans only retain 30 days of raw data. Upgrade to Pro for 1 year of raw data plus weekly and monthly aggregates that are kept indefinitely. Compare plans