---
title: "Code Review · Docs · SourceAnt"
description: "Documentation: Review local working-tree changes and pull requests against the code and context available to your SourceAnt instance."
url: "https://sourceant.ai/docs/code-review/"
---

# Code Review

Review changes in your local checkout before pushing, or review pull requests where your team works. Both use the code reviewer with the skills and context available to the instance running the review.

## Review before a pull request

Start `sourceant ui`, register your checkout, and open **Reviews**. Use **Read what changed** for an inspection without a model, or configure a model and choose **Review it** for findings. Your coding assistant can request the same review through the local MCP tool `review_working_tree`.

Follow [Local reviews](https://sourceant.ai/docs/local-reviews) for setup, review results and verification limits.

## Pull-request reviews

1.  Your repository sends PR events to SourceAnt through the GitHub App (cloud) or your own webhook configuration (self-hosted).
2.  The reviewer analyzes the diff with the configured model and posts review comments on the pull request.
3.  The review draws on the same context interfaces as the rest of the platform. Pull-request code evidence stays pinned to the commit under review, so uncommitted symbols from a local working tree cannot enter that review. Local working-tree reviews read the checkout itself. SourceAnt Cloud adds managed team context and continuous indexing at scale.

## Where reviews show up

-   **In the local browser.** Working-tree findings, skill results and review history before you push.
-   **On the pull request.** Inline comments where your team already works.
-   **In SourceAnt Lens.** A dedicated surface for reviewing changes with human-in-the-loop oversight. See [SourceAnt Lens](https://sourceant.ai/lens).

## Choose what the review includes

Open **Settings → Review** and check the selected scope before saving:

-   **Include nitpicks** is off by default. Enable it for style, naming, clarity, documentation, refactoring and optional improvement suggestions.
-   **Expert review passes** accepts `auto` for automatic selection, or exact skill identifiers, one per line. An empty list disables expert passes while general review and guidance still apply.
-   **Files excluded from review** accepts glob patterns, one per line. Lockfiles are excluded by default; dependency manifests remain included. Clear the list to review every file.
-   **Show minor suggestions in overview**, **Show critical findings in overview** and **Show regressions in overview** control those summary sections independently. All three are on by default.

See [Skills](https://sourceant.ai/docs/skills) to manage reusable instructions and [Settings](https://sourceant.ai/docs/settings#scope) for inheritance and reset behaviour.

## Running it

-   **Local:** review a checkout with your own model key. See [Local reviews](https://sourceant.ai/docs/local-reviews).
-   **Cloud:** sign up, connect your repositories through onboarding, and reviews start on your next PR. See [Cloud Setup](https://sourceant.ai/docs/cloud).
-   **Self-hosted:** run the open-source core with your own model keys. See [Self-Hosting](https://sourceant.ai/docs/self-hosting).

## Related

-   [Knowledge Management](https://sourceant.ai/docs/knowledge-management): the layer reviews draw their context from.
-   [Issue Triage](https://sourceant.ai/docs/repo-management): the other automation that runs on your PRs and issues.
