---
title: "AI Code Review with Cross-Repository Context · SourceAnt"
description: "Review pull requests with captured engineering decisions, contracts and cross-repository context. Read findings on GitHub or follow the evidence in SourceAnt Lens."
url: "https://sourceant.ai/code-review/"
---

Code review

# AI code review informed by your team’s decisions.

Review pull requests with captured engineering decisions, contracts and cross-repository context. Read findings on GitHub or follow the evidence in SourceAnt Lens.

[Start free](https://app.sourceant.ai/) [Explore the product](https://sourceant.ai/code-review/#overview)

Every engineer and every agent included, never per seat · Your own model keys

Code review

![SourceAnt reviews workspace showing pull requests and previously delivered reviews](https://sourceant.ai/_astro/reviews-dark.eoxg2H9a_1Xrqsy.webp)

Sound familiar?

## The diff cannot tell you why the code was written.

A change can look tidy and still break a deliberate constraint. Review needs the decisions around the code as well as the lines being changed.

## Check the decision behind the diff

A payment retry change can be valid within one file and still violate a recorded agreement with a mobile client. SourceAnt brings captured constraints into review so a finding can explain which agreement the change needs to preserve.

## Review changes across repository boundaries

Group related repositories into a system and record their connections. Cross-repository review follows the change into related code and reports which repositories it could not read. The result depends on the repositories and context available to your instance.

## Control how much feedback reaches your team

Nitpicks are off by default. Choose expert review skills, exclude files and control which findings appear in the overview. Review feedback on GitHub or read the risk and supporting evidence in Lens.

## Connect the context to the change.

-   [Shared memory for AI coding agents](https://sourceant.ai/knowledge-graph)
-   [AI code review with engineering context](https://sourceant.ai/code-review)
-   [Cross-repository code review and system architecture](https://sourceant.ai/systems)
-   [Connect a repository and run your first review](https://sourceant.ai/docs/getting-started)

How it works

## From the first signal to the next action.

1.  01
    
    ### Connect your repository
    
    Use the GitHub App on cloud or configure webhooks when you self-host.
    
2.  02
    
    ### Review with context
    
    SourceAnt analyzes the diff with the configured model and the context available to that instance.
    
3.  03
    
    ### Read the findings
    
    See comments on the pull request or examine the change and its evidence in Lens.
    

Built around your work

## Keep the review in your workflow.

Your team can read findings on GitHub. Use Lens when you need a dedicated view of the risk, evidence and unresolved claims.

[Read the code review documentation](https://sourceant.ai/docs/code-review)

## Before you get started.

### Where do review comments appear?

Findings appear on the pull request. Lens provides a dedicated surface for reading the change and its evidence.

### Can local work affect a remote review?

Code evidence stays pinned to the commit under review. A local review can use durable knowledge only from that same local instance.

### Can I use my own model keys?

Yes. Connect your own model provider or use cloud models billed separately from the platform. The open source core can also run on your own infrastructure.

SourceAnt · Code review

## AI code review informed by your team’s decisions.

[Start free](https://app.sourceant.ai/)[Find your plan](https://sourceant.ai/pricing)