class Riffer::Guardrail

Base class for guardrails that process input and output in the agent pipeline.

class MyGuardrail < Riffer::Guardrail
  def process_input(messages, context:)
    # Return pass(messages), transform(modified_messages), or block(reason)
    pass(messages)
  end

  def process_output(response, messages:, context:)
    # Return pass(response), transform(modified_response), or block(reason)
    pass(response)
  end
end

Public Instance Methods

name ()

Source

# File lib/riffer/guardrail.rb, line 38
def name
  Riffer::Helpers::ClassNameConverter.convert(self.class.name)
end

Returns the guardrail’s identifier, used as the tracing span suffix and the riffer.guardrail.name attribute; override to label the span.

process_input (messages, context:)

Source

# File lib/riffer/guardrail.rb, line 22
def process_input(messages, context:)
  pass(messages)
end

Processes input messages before they’re sent to the LLM; override in subclasses.

process_output (response, messages:, context:)

Source

# File lib/riffer/guardrail.rb, line 30
def process_output(response, messages:, context:)
  pass(response)
end

Processes the output response after it’s received from the LLM; override in subclasses.

Protected Instance Methods

block (reason, metadata: nil)

Source

# File lib/riffer/guardrail.rb, line 61
def block(reason, metadata: nil)
  Riffer::Guardrails::Result.block(reason, metadata: metadata)
end

Creates a block result that halts execution.

pass (data)

Source

# File lib/riffer/guardrail.rb, line 47
def pass(data)
  Riffer::Guardrails::Result.pass(data)
end

Creates a pass result that continues with unchanged data.

transform (data)

Source

# File lib/riffer/guardrail.rb, line 54
def transform(data)
  Riffer::Guardrails::Result.transform(data)
end

Creates a transform result that continues with transformed data.