The Agent Skills Directory

[SAFE]: The skill serves as a legitimate technical guide for the Himalaya CLI tool. All provided examples use standard tool features for email management via IMAP and SMTP.
[CREDENTIALS_UNSAFE]: The configuration references correctly use placeholders (e.g., "YOUR_PASSWORD", "YOUR_GOOGLE_CLIENT_ID") and promote the use of secure credential management through external commands like secret-tool or 1Password CLI instead of hardcoding sensitive data in configuration files.
[COMMAND_EXECUTION]: The skill demonstrates standard shell scripting patterns for email automation, such as using jq to parse JSON output and bash loops for bulk operations. These patterns are appropriate for the tool's intended use case and do not involve unsafe command injection or privilege escalation.

himalaya