The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to wrap the 1C Enterprise platform command-line interface. It executes local binaries (1cv8.exe and ibcmd.exe) using PowerShell and Python scripts to perform administrative tasks like listing, checking, and deleting configuration extensions.
[DYNAMIC_EXECUTION]: The scripts support an 'escape hatch' for passing additional platform arguments via -AdditionalV8Arguments and -AdditionalIbcmdArguments. To mitigate the risk of parameter injection or unauthorized command execution, the implementations include an Assert-ExtraArgs function. This function validates user input against a list of restricted/owned keys and prevents the injection of conflicting batch commands.
[CREDENTIALS_SAFE]: The skill manages sensitive credentials for database and repository access. Both the PowerShell and Python implementations incorporate a Protect-Secrets mechanism that identifies and redacts password values from console output and log files, ensuring secrets are not exposed in the agent's interaction history.
[DATA_EXPOSURE]: The skill reads .v8-project.json from the local workspace to resolve environment settings. This behavior is standard for developer tooling and is performed through controlled filesystem lookups restricted to the project directory structure.

db-cfe-admin