The Agent Skills Directory

[SAFE]: The skill provides comprehensive security guidance for network automation, explicitly advising against common pitfalls like hardcoded credentials and unbounded network scans.
[CREDENTIALS_UNSAFE]: The skill follows security best practices by instructing the user to use environment variables (os.environ.get), secure input (getpass), or a vault, and includes explicit warnings against hardcoding passwords or keys.
[INDIRECT_PROMPT_INJECTION]: While the skill processes external device output via send_command, it includes defensive measures such as structured parsing checks and advice to verify raw output, minimizing the risk of the agent misinterpreting malicious device responses as valid instructions.
[COMMAND_EXECUTION]: The skill demonstrates legitimate use of Python's netmiko and concurrent.futures for network management. It encourages safe execution patterns like dry-run flags and explicit operator approval before modifying device configurations.

netmiko-ssh-automation