The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses scripts/export_config.sh and scripts/diagnostics.py to manage configuration and perform diagnostics. These scripts perform standard operations such as reading local configuration files and making network requests to user-defined ByteHouse hosts.
[EXTERNAL_DOWNLOADS]: The skill makes network requests to the ByteHouse host provided by the user (e.g., https://{host}/matrix/v1/conversation). As this is a core function of the skill to communicate with the service it is designed for, and the destination is controlled by the user, this is considered safe and standard behavior.
[CREDENTIALS_UNSAFE]: The skill correctly instructs users to provide their host and password, which are then stored locally in ~/.bytehouse_config.json. This is a standard and acceptable practice for local CLI tools to avoid repeated credential entry, provided the user secures their local environment.
[DATA_EXFILTRATION]: While the skill transmits a password to the user-specified host via an Authorization header, this is the intended mechanism for authenticating with the ByteHouse API. No data is sent to unauthorized or hidden third-party locations.

byted-bytehouse-cluster-diagnostics