The Agent Skills Directory

[SAFE]: The skill interfaces with the official ZoomKit API through dedicated shell scripts. All network requests are directed to the vendor's primary domain (zoomkit.ru) using secure HTTPS protocols. The script includes logic to prevent data transmission over insecure HTTP or to unauthorized custom API bases unless specifically enabled by the user for local testing purposes.
[SAFE]: Credential management adheres to security best practices. The skill uses a .env configuration file for API tokens, which is excluded from version control and protected by restricted file permissions (chmod 600). Instructions explicitly warn against exposing tokens in chat, logs, or repositories.
[SAFE]: The skill implements robust input and output validation. The scripts/zoomkit.sh utility utilizes jq to verify the integrity of JSON payloads before they are sent to the API. It also provides --dry-run and --confirm mechanisms to ensure that destructive or modifying actions are reviewed by the user.
[SAFE]: No malicious patterns such as obfuscation, privilege escalation, or persistence mechanisms were identified. The use of Russian language in the scripts and documentation is legitimate and appropriate for the target audience and service (Yandex.Direct).

zoomkit