Skip to main content

Sentences Count

The Sentences count guardrail validates that the message content falls within a specified sentence count range. This is useful to ensure that prompts or responses contain a minimum amount of structured content, or to prevent overly verbose outputs.

It can be applied before sending the prompt to the LLM and after to validate the LLM response.

How it works​

The guardrail splits the message content by sentence-ending punctuation (., !, ?) and counts the resulting segments. If the count falls outside the configured min and max bounds, the message is blocked.

Configuration​

"guardrails": [
  {
    "enabled": true,
    "before": true,
    "after": true,
    "id": "sentences",
    "config": {
      "min": 1,
      "max": 50
    }
  }
]

Field explanations​

  • enabled: true — The guardrail is active
  • before: true — The guardrail applies to user input before sending to the LLM
  • after: true — The guardrail applies to the LLM response

Config section​

ParameterTypeRequiredDefaultDescription
minnumberNo0Minimum number of sentences allowed in the message
maxnumberNounlimitedMaximum number of sentences allowed in the message

Guardrail example​

If the LLM generates a response with 60 sentences and the maximum is set to 50, the response will be blocked with:

This message has been blocked by the 'sentences count' guardrail !