> ## 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.

# Security Policy • AI Agent SDK

> SecurityPolicy: Security policy for sandbox execution.

# SecurityPolicy

> Defined in the [**config**](../modules/config) module.

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

Security policy for sandbox execution.

Attributes:
allow\_network: Whether to allow network access
allow\_file\_write: Whether to allow file writes
allow\_subprocess: Whether to allow subprocess creation
allowed\_paths: List of paths that can be accessed
blocked\_paths: List of paths that are blocked
allowed\_commands: List of allowed shell commands (empty = all)
blocked\_commands: List of blocked shell commands
allowed\_imports: List of allowed Python imports (empty = all)
blocked\_imports: List of blocked Python imports
max\_output\_size: Maximum output size in bytes

## Properties

<ResponseField name="allow_network" type="bool">
  No description available.
</ResponseField>

<ResponseField name="allow_file_write" type="bool">
  No description available.
</ResponseField>

<ResponseField name="allow_subprocess" type="bool">
  No description available.
</ResponseField>

<ResponseField name="allowed_paths" type="List">
  No description available.
</ResponseField>

<ResponseField name="blocked_paths" type="List">
  No description available.
</ResponseField>

<ResponseField name="allowed_commands" type="List">
  No description available.
</ResponseField>

<ResponseField name="blocked_commands" type="List">
  No description available.
</ResponseField>

<ResponseField name="allowed_imports" type="List">
  No description available.
</ResponseField>

<ResponseField name="blocked_imports" type="List">
  No description available.
</ResponseField>

<ResponseField name="max_output_size" type="int">
  No description available.
</ResponseField>

## Methods

<CardGroup cols={2}>
  <Card title="strict()" icon="function" href="../functions/SecurityPolicy-strict">
    Create a strict security policy for untrusted code.
  </Card>

  <Card title="standard()" icon="function" href="../functions/SecurityPolicy-standard">
    Create a standard security policy.
  </Card>

  <Card title="permissive()" icon="function" href="../functions/SecurityPolicy-permissive">
    Create a permissive security policy for trusted code.
  </Card>
</CardGroup>

<Accordion title="Internal & Generic Methods">
  * **to\_dict**: Convert to dictionary.
</Accordion>

## Source

<Card title="View on GitHub" icon="github" href="https://github.com/MervinPraison/PraisonAI/blob/main/src/praisonai-agents/praisonaiagents/sandbox/config.py#L14">
  `praisonaiagents/sandbox/config.py` at line 14
</Card>