create-skill

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions for environment setup using standard shell commands (e.g., mkdir -p) to create the necessary directory structure for new skills. It also defines how to reference and execute scripts within the scripts/ directory for deterministic tasks.
  • [PROMPT_INJECTION]: As a skill designed to generate instructions for other skills, it possesses an inherent surface for indirect prompt injection.
  • Ingestion points: User-provided inputs for skill names, descriptions, and procedural workflows in SKILL.md (Steps 1, 3, and 4).
  • Boundary markers: None explicitly required for the generated content beyond YAML frontmatter delimiters.
  • Capability inventory: The skill facilitates the use of Write, Edit, and Bash tools to create and modify executable-capable skill files.
  • Sanitization: No explicit sanitization of user-provided content is mentioned before it is written to the file system. However, the skill encourages objective, imperative language which serves as a natural defensive instruction style.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 08:48 AM