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

# Correctness configuration

> Triggers, labels, severity thresholds, and scope for the built-in Correctness check.

Macroscope's Correctness check reviews pull requests for bugs and posts findings on GitHub. In GitHub, you can reply to a finding with context, give feedback, or use [Fix it for me](/fix-it-for-me) to request a change.

<a id="detection-mode" />

<a id="configuring-detection-mode" />

<a id="adjusting-detection-modes" />

Choose a [Detection mode](/code-review/detection-mode) to control coverage, validation, and cost. For custom checks with your own instructions, model, and tools, use [Check Run Agents](/check-run-agents).

Configure repository settings in **Settings → Repos**. Personal settings override repository preferences for the PRs you author. **Defaults** seeds newly linked repositories; it does not update existing ones.

## Review triggers

In **Settings → Repos**, set **Correctness** to **Always run**, **Manual invocation only**, or **Never run**. **Always run** reviews PRs when they open or receive pushes. **Manual invocation only** reviews only when someone mentions `@macroscope-app` or adds an **Always Review PR Label**. **Never run** turns it off completely, and nothing overrides it.

Onboarding turns Correctness on for the repositories you pick. Repositories linked later follow **Defaults**, which start off.

<a id="code-review-settings" />

| Repository control | Effect |
| - | - |
| Review Draft PRs | Review before the PR is ready; the author's personal preference can override this.<br /><br />Switching a draft PR to ready for review will trigger a Macroscope Correctness check. |
| Review Cross-Repo PRs | Allow reviews of fork and external PRs. |
| Skip Dependabot | Skip Dependabot PRs. |
| Skip PRs by Author / Labels / Target Branches | Exclude matching PRs. Write bot authors with the `[bot]` suffix, as in `my-bot[bot]`. |
| Always Review PR Labels | Run checks even when automatic review is off. |
| Max Automatic Runs per PR | Skip later automatic correctness reviews after the cap, titled "Maximum runs reached". Manual requests, check reruns, and draft-to-ready transitions are exempt. |
| Max Reviewable File Size | Exclude files above the configured size in KB, listed as "File too large (exceeds configured limit)". A custom correctness instruction's explicit `include` overrides this. |

The run-count and file-size limits are optional; blank or `0` clears them. Without a configured size limit, files of any size are reviewed.

## PR labels

| Label type | Behavior |
| - | - |
| Skip label | Prevents automatic review; wins over ordinary always-review labels. |
| Always-review label | Enables review when automatic review is off. |
| `macroscope-budget`, `macroscope-balanced`, `macroscope-precise`, `macroscope-ultra` | Selects that mode until removed; also triggers correctness, Check Run Agents, and Approvability. Overrides repository skip-label and target-branch rules, but not skip authors. |

Two labels naming different modes prevent reviews from running. Removing a mode label does not itself trigger a review. See [Check Run Agent trigger rules](/check-run-agents/triggers) for personal overrides and per-agent filters.

## Minimum severity to comment

Set **Minimum Severity to Comment** to **Low**, **Medium**, **High**, **Critical**, or **Off** (default). Off posts all findings. Other values withhold findings below the threshold and findings without recorded severity.

The PR author's personal threshold overrides the repository threshold, including in CLI reviews. This affects correctness comments only. Withheld findings remain in the check's **Filtered Issues Details**, tagged **Below severity threshold**, and remain available to Approvability.

[Minimum Blocking Severity](/approvability#minimum-blocking-severity) is separate: it controls which outstanding correctness comments block automatic approval.

### Issue severity levels

| Severity | Examples |
| - | - |
| Critical | Data loss or security breach. |
| High | Production crash or security degradation. |
| Medium | Broken but recoverable functionality. |
| Low | Cosmetic or edge-case issues. |

## Manually triggering a review

```text theme={null}
@macroscope-app review
```

A manual request **bypasses** automatic run-count and correctness spend caps. It does not bypass the workspace balance, monthly spend limit, ignore rules, or file-size rules. See [spend controls](/spend-limits).

<Tip>
  Unless a Detection Mode is [explicitly mentioned](/code-review/detection-mode#choose-a-mode) (or the author has a personal override configured), the review will use the repository’s configured Detection Mode.
</Tip>

## Ignore rules and custom instructions

<a id="excluding-files-with-macroscopeignoremd" />

<a id="excluding-files-from-code-review" />

<a id="macroscope-ignore" />

<a id="default-ignore-patterns" />

<a id="reviewing-your-test-code" />

* [Ignore rules](/code-review/ignore-files): exclude generated files, dependencies, or other paths; opt into reviewing tests with `ignoreTests: false`.
* [Custom instructions](/custom-instructions): supply domain rules scoped to particular files, or wait for CI before reviewing.

## Results

<a id="code-review-activity-log" />

<a id="how-reviews-appear-in-github" />

Correctness reports `neutral` when it finds issues. Review details list excluded files and why they were skipped. New pushes re-evaluate findings; resolved comments link to the commit that addressed them.

The app's Code Review activity log filters by source, including Correctness and Check Run Agents. Filters persist in the URL and apply to CSV exports. Feedback metrics can also be filtered by review type.

## Language support

<a id="how-does-code-review-work" />

<a id="how-code-review-works" />

Correctness reviews every language, plus configuration, documentation, and scripts, subject to file exclusions. It explores your codebase through its syntax tree and can use web search for current library and API documentation.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.