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

# validator • AI Agent SDK

> Skill validation logic.

# validator

<Badge color="blue">AI Agent</Badge>

Skill validation logic.

## Import

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
from praisonaiagents.skills import validator
```

## Functions

<CardGroup cols={2}>
  <Card title="validate_metadata()" icon="function" href="../functions/validate_metadata">
    Validate parsed skill metadata.
  </Card>

  <Card title="validate()" icon="function" href="../functions/validate">
    Validate a skill directory.
  </Card>
</CardGroup>

### Constants

| Name                       | Value                                                                                                                                                                                                         |
| -------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `MAX_SKILL_NAME_LENGTH`    | `64`                                                                                                                                                                                                          |
| `MAX_DESCRIPTION_LENGTH`   | `1024`                                                                                                                                                                                                        |
| `MAX_COMPATIBILITY_LENGTH` | `500`                                                                                                                                                                                                         |
| `ALLOWED_FIELDS`           | `{'name', 'description', 'license', 'allowed-tools', 'metadata', 'compatibility', 'when_to_use', 'disable-model-invocation', 'user-invocable', 'argument-hint', 'model', 'effort', 'context', 'agent', 'h...` |