Skip to main content

Prompt contains gibberish guardrail

This guardrail acts like a filter that detects and manages inputs that are nonsensical, random, or meaningless, preventing the AI from generating irrelevant or low-quality responses. It uses a dedicated LLM provider with a hardcoded prompt to detect gibberish content.

This can be applied before the LLM processes the input (blocking nonsense prompts) and after to filter out meaningless responses.

How it works​

The guardrail sends messages to a validation LLM with a system prompt that instructs it to determine whether the content is gibberish. The validation LLM responds with true (content is meaningful) or false (content is gibberish).

Configuration​

"guardrails": [
  {
    "enabled": true,
    "before": true,
    "after": true,
    "id": "gibberish",
    "config": {
      "provider": "provider_xxxxxxxxx"
    }
  }
]

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
providerstringYes—Reference ID of the LLM provider used to evaluate messages for gibberish. Must be different from the main provider, unless a model is set.
modelstringNodefault model of the providerModel of the validation provider used to judge the messages, for instance a smaller and faster one. With a model set, the guarded provider can judge its own messages with another of its models.
err_msgstringNo"This message has been blocked by the 'gibberish' guardrail !"Custom error message returned when a message is blocked.

Guardrail example​

If a user inputs, "asdkjfhasd lkj3r2lkkj!!", the LLM will recognize it as gibberish and block the request.

If the LLM generates a nonsensical response, it will be flagged and removed.