The Agent Skills Directory

[PROMPT_INJECTION]: The skill acts as an ingestion surface for untrusted external data, as it is designed to process and rewrite user-provided Korean text from files or conversation context.
Ingestion points: Text provided via arguments, dialogue, or file paths (Step 1 in SKILL.md).
Boundary markers: None explicitly defined to separate user text from agent instructions.
Capability inventory: Uses Read, Write, Edit, Grep, and Glob tools to analyze and modify files.
Sanitization: No specific input sanitization or filtering logic is present. While this creates a surface for indirect prompt injection, it is the primary intended purpose of the skill and does not contain malicious logic or bypass instructions.

korean-humanizer