Word Count Guardrail¶

The Word Count Guardrail is a custom Synapse mediator for the WSO2 API Manager Universal Gateway that enforces minimum and/or maximum word count limits on a selected JSON field in AI API requests or GenAI responses. It helps control verbosity, prompt injection surface, and output consistency.

Features¶

Validate payload content by counting words.
Enforces configurable minimum and maximum sentence thresholds.
Target specific fields in JSON payloads using JSON Path
Optionally invert validation logic (e.g., allow only content outside the specified sentence range)
Provides an optional Show Guardrail Assessment feature to include detailed validation results in error responses for enhanced observability.

How to Use¶

Follow these steps to integrate the Word Count Guardrail policy into your AI API:

Open the API Publisher Portal (https://<host>:<port>/publisher)
Select your AI API
Go to Develop > API Configurations > Policies
Expand Common Policies in the Policy List
Drag and drop the Word Count Guardrail policy into your desired mediation flow
Fill in the required policy configuration (name, minimum and maximum word count)
Save and Deploy the AI API

Example Policy Configuration¶

Click to expand configuration steps

Example: Enforce prompts to contain between 2 and 5 words.

Create an AI API using Mistral AI.
Add the Word Count Guardrail policy to the API with the following configuration:

Field	Example
`Guardrail Name`	`Word Limiter`
`Minimum Word Count`	`2`
`Maximum Word Count`	`5`
`JSON Path`	`$.messages[-1].content`
`Invert the Guardrail Decision`	`false`
`Show Guardrail Assessment`	`true`

Save and re-deploy the API.
Invoke the API's chat/completion endpoint with a prompt that violates the word count, such as having only one word (if min is set to 2):

Test Request:

{
    "model": "mistral-small-latest",
    "messages": [
        {
            "role": "user",
            "content": "This is a short prompt."
        }
    ]
}

The following guardrail error response will be returned with http status code 446:

{
    "code": "900514",
    "type": "WORD_COUNT_GUARDRAIL",
    "message": {
        "interveningGuardrail": "Word Limiter",
        "action": "GUARDRAIL_INTERVENED",
        "actionReason": "Violation of applied word count constraints detected.",
        "direction": "REQUEST"
    }
}

Limitations¶

The Word Count Guardrail uses the following regular expression to split words from the inspected content:

\s+

This pattern splits words based on one or more whitespace characters. This approach may not accurately handle cases with punctuation attached to words, special characters, or languages without clear whitespace delimiters.