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

# Enterprise Features

> Audit trails, compliance, and enterprise-grade capabilities

# Enterprise Features

For organizations with compliance requirements, high-volume needs, or
enterprise-grade security expectations, mixus Legal AI provides
comprehensive features to meet your standards.

## Complete Audit Trail

Every document processed includes a full audit trail:

<Frame>
  ```markdown theme={null}
  # Audit Trail Report

  **Audit ID:** audit_1705312200_x7k9m2
  **Document Type:** NDA
  **Original File:** Acme-NDA-v1.docx
  **Status:** Success
  **Processing Time:** 12,450ms

  ## Timeline

  | Time         | Action                      | Details                     |
  | ------------ | --------------------------- | --------------------------- |
  | 10:30:00.000 | audit_started               | userId: user_123            |
  | 10:30:00.150 | classification_completed    | type: nda, confidence: 0.95 |
  | 10:30:02.300 | edit_attempt                | attempt: 1, maxRetries: 3   |
  | 10:30:05.800 | validation_completed        | passed: true                |
  | 10:30:08.200 | output_generation_completed | files: 3                    |
  | 10:30:12.450 | audit_finished              | status: success             |

  ```
</Frame>

### What's Tracked

<CardGroup cols={2}>
  <Card title="Every Operation" icon="list-check">
    Classification, editing, validation, output generation—all logged.
  </Card>

  <Card title="Timestamps" icon="clock">
    Precise timing for each step of the process.
  </Card>

  <Card title="Model Versions" icon="code-branch">
    Which AI models were used and their versions.
  </Card>

  <Card title="User Attribution" icon="user">
    Who initiated the operation and from which organization.
  </Card>
</CardGroup>

***

## Validation Tiers

Choose the validation level appropriate for your needs:

### Standard Tier

Best for: General documents, quick edits

<Check>Content accuracy check</Check>
<Check>Format integrity validation</Check>
<Check>\~5,000 tokens used</Check>

### Legal Tier

Best for: Contracts, legal documents

<Check>All standard tier checks</Check>
<Check>Risk analysis</Check>
<Check>Section compliance</Check>
<Check>Visual verification</Check>
<Check>\~15,000 tokens used</Check>

### Enterprise Tier

Best for: High-stakes documents, compliance-critical work

<Check>All legal tier checks</Check>
<Check>Human review flag</Check>
<Check>Extended audit logging</Check>
<Check>\~25,000 tokens used</Check>

***

## Human Review Integration

Enterprise tier includes human review flags:

```markdown theme={null}
⚠️ HUMAN REVIEW REQUIRED

Document: Acme-Enterprise-MSA.docx
Reason: High-risk changes detected

Flagged Items:
1. Indemnification scope expanded (Section 8.1)
2. Liability cap removed (Section 9.2)

Action Required: Senior counsel review before delivery
```

<Note>
  Human review flags don't block delivery—they ensure appropriate eyes see the
  document before final approval.
</Note>

***

## Security & Compliance

### Data Handling

<CardGroup cols={2}>
  <Card title="SOC 2 Type II" icon="shield-check">
    Independently audited security controls
  </Card>

  <Card title="Encryption" icon="lock">
    AES-256 encryption at rest, TLS 1.3 in transit
  </Card>

  <Card title="Data Isolation" icon="box">
    Organization data is strictly isolated
  </Card>

  <Card title="Retention Control" icon="clock-rotate-left">
    Configurable data retention policies
  </Card>
</CardGroup>

### Where Data Lives

* **Document Storage**: AWS S3 with server-side encryption
* **Processing**: Temporary buffers cleared after processing
* **Audit Logs**: Retained per your organization's policy
* **No Training**: Your documents are never used for model training

***

## High-Volume Processing

For organizations processing many documents:

### Parallel Processing

Multiple documents can be processed simultaneously without degradation in
quality or speed.

### Rate Limiting

We implement fair rate limits to ensure consistent performance:

| Tier       | API Calls | Per Hour            |
| ---------- | --------- | ------------------- |
| Standard   | 50        | Document operations |
| Enterprise | Custom    | Based on agreement  |

### Bulk Operations

Coming soon: API endpoints for bulk document processing with batch
reporting and consolidated audit trails.

***

## Integration Options

### API Access

Programmatic access to all Legal AI capabilities:

```bash theme={null}
POST /api/legal-ai/redline
{
  "documentKey": "s3://bucket/document.docx",
  "instructions": "Change governing law to Texas",
  "validationTier": "legal",
  "maxRetries": 3
}
```

### Webhooks

Receive notifications when processing completes:

```json theme={null}
{
  "event": "redline.completed",
  "auditId": "audit_1705312200_x7k9m2",
  "status": "success",
  "files": {
    "redlinedDocx": "s3://...",
    "cleanDocx": "s3://...",
    "comparisonPdf": "s3://..."
  }
}
```

### SSO Integration

Enterprise single sign-on support:

* SAML 2.0
* OpenID Connect
* Active Directory Federation

***

## Reporting & Analytics

Track usage and quality metrics:

### Usage Reports

* Documents processed by type
* Validation pass/fail rates
* Average processing time
* Token usage and costs

### Quality Metrics

* Changes by category (requested/derived/flagged)
* Risk flags by type
* Retry frequency
* Validation failure reasons

***

## Getting Started with Enterprise

<Card title="Contact Sales" icon="envelope" href="mailto:sales@mixus.ai">
  Discuss enterprise features, custom requirements, and pricing.
</Card>

<Card title="Security Documentation" icon="file-shield" href="/security">
  Review our security practices and compliance certifications.
</Card>