The Agent Skills Directory

[SAFE]: The skill interacts with official Volcengine services using the volcengine-python-sdk. All network operations are targeted at well-known vendor endpoints for RDS instance management.
[COMMAND_EXECUTION]: The skill utilizes a local helper script scripts/call_rds_mssql.py to perform specific management actions like list-instances and describe-db-accounts. This execution is restricted to predefined subcommands and is essential for the skill's primary function.
[SAFE]: Sensitive credentials such as access keys and secret keys are managed through environment variables rather than being hardcoded, which aligns with secure development standards.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to interpret natural language instructions to drive cloud management tasks. 1. Ingestion points: User natural language requests interpreted by the agent as defined in SKILL.md. 2. Boundary markers: The instructions provide context for intent recognition but do not implement specific data delimiters. 3. Capability inventory: The skill can invoke the call_rds_mssql.py script to make authenticated API calls to Volcengine cloud resources. 4. Sanitization: Parameters passed to the management script are validated using the argparse library, providing a layer of protection against command injection.

byted-sqlserver-instance-management