The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to use standard Git commands such as git status, git stash list, and git stash push. These are routine developer operations and are used in a controlled manner.
[CREDENTIALS_UNSAFE]: The instructions explicitly include a security best practice: when stashing changes, the agent is directed to use --include-untracked rather than --all to ensure that ignored files (which may contain credentials or secrets) are not accidentally moved into the stash.
[REMOTE_CODE_EXECUTION]: No remote code execution or external package installations were detected. The skill relies on local environment tools.
[DATA_EXFILTRATION]: There are no network operations or patterns suggesting data is being sent to external servers.
[PROMPT_INJECTION]: The instructions provide guidance on agent behavior and troubleshooting without attempting to bypass safety filters or override fundamental system constraints.

hl-checkpoint