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

# /analyze

> Comprehensive code analysis across quality, security, performance, and architecture.

## What it does

`/analyze` runs a comprehensive review across four domains: code quality, security, performance, and architecture. You can target a specific domain or run all four for a full report. Each finding is reported with a severity level and a specific fix recommendation.

## When to use

Use `/analyze` when you want an objective assessment of your codebase — before a major refactor, after a sprint, when onboarding to an existing codebase, or as a pre-ship checklist. For a focused security review run with `--focus security`; for a performance investigation use `--focus performance`.

## Prerequisites

* The code you want to analyze is present in the working directory

## Conversation mode

Either mode works.

## What happens

<Steps>
  <Step title="Quality analysis">
    Code smells, duplication, complexity, maintainability index, dead code, unused imports, and naming conventions are reviewed.
  </Step>

  <Step title="Security analysis">
    The `security-review` skill is loaded for the full OWASP checklist — auth, input handling, secrets exposure, and sensitive data handling.
  </Step>

  <Step title="Performance analysis">
    The `performance-optimization` skill is loaded. Performance is measured before analyzing — no recommendations without data.
  </Step>

  <Step title="Architecture analysis">
    The `architecture-design` skill is loaded for structural analysis — component relationships, boundaries, coupling, and cohesion.
  </Step>

  <Step title="Report findings">
    Each finding is reported with: domain, severity, location (file:line), description, and a specific fix recommendation.
  </Step>

  <Step title="Prioritize actions">
    Critical and High findings must be fixed before proceeding. Medium findings become tasks. Low and Info findings are documented for future improvement.
  </Step>
</Steps>

## Skills invoked

* `security-review` — OWASP checklist for security domain
* `performance-optimization` — measurement-first performance analysis
* `architecture-design` — structural and coupling analysis

## Usage

```bash theme={null}
/analyze
/analyze --focus quality
/analyze --focus security
/analyze --focus performance
/analyze --focus architecture
```

## Output format

For each finding:

| Field          | Description                                     |
| -------------- | ----------------------------------------------- |
| Domain         | quality / security / performance / architecture |
| Severity       | critical / high / medium / low / info           |
| Location       | file:line                                       |
| Issue          | What's wrong                                    |
| Recommendation | Specific fix                                    |

## Example

```bash theme={null}
/analyze --focus security
```

Antigravity loads the `security-review` skill and reports findings like:

```
Domain: security
Severity: HIGH
Location: src/api/users.js:47
Issue: User input from req.body.username passed directly to SQL query — SQL injection risk
Recommendation: Use parameterized queries: db.query('SELECT * FROM users WHERE id = ?', [req.body.username])
```

## Related commands

<CardGroup cols={2}>
  <Card title="/improve" href="/commands/improve">
    Apply the fixes that /analyze identifies.
  </Card>

  <Card title="/review" href="/commands/review">
    Code review against plan compliance — complements /analyze's technical focus.
  </Card>

  <Card title="/troubleshoot" href="/commands/troubleshoot">
    For issues that /analyze surfaces that require deep debugging.
  </Card>

  <Card title="/design" href="/commands/design">
    Address architectural issues that /analyze flags by redesigning the affected components.
  </Card>
</CardGroup>