# AI code review that runs your code.

Test Run EXecute

**TREX** runs your PR branch in a sandbox to find the bugs that only show up at runtime. It starts services, mocks inputs, clicks through UI flows, and leaves the logs, screenshots, and traces on the PR.

[Contact sales](/contact?source=trex)

[Get started](https://app.greptile.com/signup)

\[ TREX \]

## How TREX validates a PR

### How it works

#### 01

##### Understands the change

TREX looks at the PR, your repo, your test suite, and your stack to figure out what needs to run.

#### 02

##### Runs it in a sandbox

TREX spins up the tools it needs: services, dev servers, mocks, API calls, browser agents, or targeted tests.

#### 03

##### Shows what broke

When something fails, TREX traces it back to the relevant code and leaves a PR comment along with the evidence.

\[ TREX \]

## Get the bugs you can't find by reading the diff

Greptile with TREX catches ~20% more bugs than review alone. Those are the bugs that only show up when the code runs.

### Test every change end to end

TREX spins up the services your change touches and uses computer use to test them. It clicks through desktop apps, web apps and command-line tools the way a real user would.

[Get started](https://app.greptile.com/signup)

\[ MORE FEATURES \]

## Explore other features at Greptile

Find out more about how teams are using Greptile

[See docs](https://www.greptile.com/docs/introduction)

### [Agent](/agent.md)

The code reviewer that catches real bugs

[Learn more](/agent.md)

### [Independence](/independence.md)

The independent code validator

[Learn more](/independence.md)

### [Learning & Custom Context](/learning.md)

Code reviews that get personalized over time

[Learn more](/learning.md)

## Let Greptile run the PR before it merges.

[View Pricing](/pricing.md)

[Start now](https://app.greptile.com/signup)