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

# Reports

> The readiness score, per-turn grades, and failures from a run.

A **report** is the output of a test run — a shareable document that shows how ready your agent is to ship.

## What's in a report

| Section          | Description                                                          |
| ---------------- | -------------------------------------------------------------------- |
| Readiness Score  | An overall 0–100 score for the run                                   |
| Per-turn grades  | Accuracy, refusal, and consistency for every conversation turn       |
| Failures         | Turns that fell below the passing threshold, with the exact exchange |
| Red-team results | Whether your agent resisted adversarial attacks                      |

## Sharing

Every report has a public link you can share with stakeholders, paste into a PR, or attach to a release. Reports are versioned — re-running produces a new report you can diff against the previous one.