InputSanitizationMiddleware

org.llm4s.llmconnect.middleware.InputSanitizationMiddleware

class InputSanitizationMiddleware(maxTotalCharacters: Int, forbiddenPatterns: Seq[Regex]) extends LLMMiddleware

Middleware that sanitizes and validates input prompts.

Can enforce max length limits and reject inputs containing forbidden patterns.

Attributes

Human-readable name for logging/debugging.

Wrap the given LLMClient, returning a new client with added behavior.

Implementations should delegate all LLMClient methods to next, adding behavior before/after delegation as needed.

In this article

Generated with