Skip to main content

​
guardrails

Rust AI Agent SDK Guardrails Module This module provides guardrail validation for agent outputs:
  • Guardrail - Guardrail trait for validation
  • GuardrailResult - Result of guardrail validation
  • LlmGuardrail - LLM-based guardrail implementation

​
Example

use praisonai::guardrails::{Guardrail, GuardrailResult};

struct ContentFilter;

impl Guardrail for ContentFilter {
fn validate(&self, output: &str) -> GuardrailResult {
if output.contains("unsafe") {
GuardrailResult::failure("Content contains unsafe material")
} else {
GuardrailResult::success(output.to_string())
}
}
}

​
Import

use praisonai::guardrails::*;

​
Classes

GuardrailResult

Result of a guardrail validation.

GuardrailConfig

Configuration for guardrails.

LengthGuardrail

Content length guardrail.

BlocklistGuardrail

Keyword blocklist guardrail.

PatternGuardrail

Regex pattern guardrail.

GuardrailChain

Chain of guardrails to run in sequence.

Guardrail

Trait for synchronous guardrail validation.

AsyncGuardrail

Trait for asynchronous guardrail validation.

GuardrailAction

Action to take when guardrail fails.

Rust Guardrails

Rust Approval

Rust Security