The Agent Skills Directory

[SAFE]: The skill operates as a guide for development tasks and does not contain malicious code, data exfiltration patterns, or obfuscated content.
[COMMAND_EXECUTION]: The skill documentation includes standard shell commands for scaffolding projects (mkdir, touch) and references the platform's native CLI tool (cc) for testing. These commands are typical for development environments and are aligned with the skill's primary purpose.
[PROMPT_INJECTION]: The skill contains instructions that interpolate user-provided parameters (such as skill names) into shell commands, creating a surface for indirect prompt injection.
Ingestion points: The file SKILL.md identifies user input for skill names and descriptions during the creation process.
Boundary markers: No specific boundary markers or delimiters are defined for the user-supplied content.
Capability inventory: The file SKILL.md includes instructions for mkdir and touch commands for directory and file creation.
Sanitization: No explicit sanitization or validation of the user-provided strings is mentioned. Given the skill's role as a developer utility, this is evaluated as a functional requirement rather than a malicious vector.

Skill Development