> ## Documentation Index
> Fetch the complete documentation index at: https://docs.pylar.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Evals Dashboard

> Navigate the Evals dashboard and understand evaluation metrics, visual insights, and filters

## Opening the Dashboard

To access the Evaluation Dashboard:

1. Navigate to your project in Pylar
2. Click the **"Eval"** button in the top-right corner of the screen
3. The Evaluation Dashboard opens

You'll see a comprehensive view of your MCP tool performance metrics.

## Dashboard Overview

The Evaluation Dashboard is organized into several sections:

* **Filters**: Select which MCP tool to review
* **Summary Metrics**: High-level performance indicators
* **Visual Insights**: Time-series graphs showing trends
* **Error Analysis**: Detailed error breakdown
* **Raw Logs**: Complete records of all tool calls

## Filters

### Selecting a Tool

At the top of the dashboard, you'll find filters to select which MCP tool you want to review.

**How to use**:

1. Click the filter dropdown
2. Select the MCP tool you want to analyze
3. The dashboard updates to show metrics for that tool only

<Tip>
  Use filters to focus on specific tools. This is especially useful when you have multiple tools and want to analyze them individually.
</Tip>

## Evaluation Metrics

The dashboard displays key metrics that summarize tool performance:

### Total Count

**What it is**: The total number of times the selected MCP tool was invoked.

**What it tells you**: Overall usage volume—how frequently agents are using this tool.

<Info>
  Total Count includes both successful and failed invocations. It's the baseline for all other metrics.
</Info>

### Success Count

**What it is**: How many invocations returned a valid result.

**What it tells you**: The absolute number of successful tool calls. Higher is better.

### Error Count

**What it is**: How many invocations failed to return a result.

**What it tells you**: The absolute number of failed tool calls. Lower is better.

### Success Rate

**Calculation**:

```
Success Rate = (Success Count ÷ Total Count) × 100
```

**What it is**: Percentage of successful tool invocations.

**What it tells you**:

* High success rate (90%+) = Tool is working well
* Medium success rate (70-90%) = Some issues, needs attention
* Low success rate (less than 70%) = Significant problems, needs immediate attention

<Check>
  Aim for success rates above 90%. If your success rate is below this threshold, investigate errors to understand what's going wrong.
</Check>

### Error Rate

**Calculation**:

```
Error Rate = (Error Count ÷ Total Count) × 100
```

**What it is**: Percentage of failed tool invocations.

**What it tells you**:

* Low error rate (less than 10%) = Tool is reliable
* Medium error rate (10-30%) = Some reliability issues
* High error rate (greater than 30%) = Major problems, needs fixing

<Warning>
  High error rates indicate problems that are affecting agent performance. Address these issues promptly to improve agent experience.
</Warning>

## Visual Insights

The dashboard includes time-series graphs that show how metrics change over time.

### Calls/Success/Errors Graph

**What it shows**: A time-series plot displaying:

* **Total Calls**: How many times the tool was invoked over time
* **Successes**: Successful invocations over time
* **Errors**: Failed invocations over time

**What it tells you**:

* Usage trends (increasing/decreasing usage)
* Performance trends (improving/declining success rates)
* Error patterns (when errors occur most frequently)

**How to use it**:

* Look for trends: Are errors increasing or decreasing?
* Identify patterns: Do errors spike at certain times?
* Compare periods: How does current performance compare to past performance?

### Success/Error Rate (%) Graph

**What it shows**: Displays success and error percentages as a time-series trend.

**What it tells you**:

* Performance stability over time
* Whether your tool is improving or degrading
* Correlation between changes and performance

**How to use it**:

* Monitor trends: Is success rate improving?
* Spot anomalies: Are there sudden drops in performance?
* Track improvements: Did your changes improve performance?

<Tip>
  Use these graphs to understand not just current performance, but also trends and patterns. This helps you identify issues before they become critical.
</Tip>

## Interpreting the Metrics

### Healthy Tool Performance

A healthy tool shows:

* ✅ High success rate (90%+)
* ✅ Low error rate (less than 10%)
* ✅ Stable or improving trends over time
* ✅ Consistent performance across time periods

### Tool Needs Attention

A tool that needs attention shows:

* ⚠️ Success rate below 90%
* ⚠️ Error rate above 10%
* ⚠️ Declining trends in graphs
* ⚠️ Inconsistent performance

### Tool Needs Immediate Fix

A tool that needs immediate attention shows:

* ❌ Success rate below 70%
* ❌ Error rate above 30%
* ❌ Sharp drops in performance graphs
* ❌ Frequent errors in logs

## Using Filters Effectively

### Analyzing Individual Tools

1. Select a specific tool from the filter
2. Review its metrics
3. Check if it meets performance thresholds
4. Investigate if it needs improvement

### Comparing Tools

1. Select different tools one at a time
2. Compare their metrics
3. Identify which tools perform best
4. Learn from high-performing tools to improve others

### Time Period Analysis

Use time filters (if available) to:

* Compare different time periods
* See impact of tool changes
* Identify seasonal patterns
* Track improvement over time

## Next Steps

Now that you understand the dashboard:

* [Analyzing Errors](/learn/evals/analyzing-errors) - Dive deeper into error patterns
* [Understanding Query Shapes](/learn/evals/understanding-query-shapes) - Learn about query patterns
* [Raw Logs](/learn/evals/understanding-query-logs) - Explore detailed execution logs

<Card title="Analyze Errors" icon="triangle-exclamation" href="/learn/evals/analyzing-errors">
  Learn how to identify and fix errors in your tools
</Card>