The Agent Skills Directory

[SAFE]: The skill is composed of markdown-based instructions, writing patterns, and examples. It does not include any scripts, executable code, or system-level commands.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to process user-provided text for humanization. This creates a surface for indirect prompt injection, but since the skill has no access to sensitive capabilities like the file system, network, or shell, the risk is minimal and contained to the generated output text. Ingestion points: User-provided text for humanization as defined in the 'Your Task' section of SKILL.md. Boundary markers: None explicitly defined in instructions for delimiting input text. Capability inventory: No file writes, network calls, or subprocess execution patterns are present across the skill files. Sanitization: No specific sanitization or filtering of external content is defined in the humanization process.

humanizer