token-smuggling

What problem does it solve?

This Skill helps detect and prevent token-smuggling and Unicode injection attacks in code that passes user input to LLMs, ensuring prompts cannot be manipulated through hidden characters or homoglyphs.

Core Features & Use Cases

• Detects Unicode control and invisible formatting characters that can be used to tamper with prompts.
• Enforces Unicode normalization and safety checks at the trust boundary to prevent bypass of filters.
• Provides guidance for secure prompt construction and verification in software that integrates LLMs.

Quick Start

Normalize all user input with NFKC before it reaches any LLM prompt construction.