create-agent-skills

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell commands to manage files within the ~/.claude/ directory, including listing, reading, and creating skill structures. This is core functionality for its purpose of skill management.
  • [COMMAND_EXECUTION]: Employs the chmod +x command to make supporting scripts executable, which is a standard procedure for skill development.
  • [EXTERNAL_DOWNLOADS]: Contains instructions and examples for installing Python packages using pip, though it does not automatically execute installations from untrusted sources.
  • [CREDENTIALS_UNSAFE]: Interacts with ~/.claude/.env for managing API credentials. Critically, it includes a comprehensive security guide (references/api-security.md) that instructs users on how to avoid exposing these credentials in chat logs by using wrapper scripts and environment variables.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 08:11 PM