> ## Documentation Index
> Fetch the complete documentation index at: https://docs.2027.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Scoring

> How eval scores and grades are calculated

# Scoring

## Grade scale

| Grade | Score Range | Meaning                                   |
| ----- | ----------- | ----------------------------------------- |
| A+    | 95-100      | Exceptional agent experience              |
| A     | 90-94       | Excellent — minimal friction              |
| A-    | 85-89       | Very good with minor issues               |
| B+    | 80-84       | Good overall experience                   |
| B     | 75-79       | Solid but with notable friction           |
| B-    | 70-74       | Acceptable with clear improvements needed |
| C+    | 65-69       | Below average — significant friction      |
| C     | 60-64       | Poor agent experience                     |
| C-    | 55-59       | Major issues throughout                   |
| D     | 40-54       | Failing — agent struggled significantly   |
| F     | 0-39        | Agent could not complete the task         |

## How scores are calculated

The overall score is a weighted average of dimension scores. Each dimension is scored independently based on the agent's experience in that area.

Scores can be adjusted by admins if the automated scoring doesn't accurately reflect the experience (e.g., excluding irrelevant friction points or accounting for external factors).