The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard Git commands and local scripts to automate the creation of project checkpoints. These commands are executed within the user's project directory and are consistent with the skill's stated purpose.
[DATA_EXPOSURE]: The profile_store.py script manages local configuration files and includes an explicit filtering mechanism to prevent sensitive information, such as passwords or API keys, from being stored in the shared user profile.
[INDIRECT_PROMPT_INJECTION]: The skill ingests Git history and project documentation as part of its workflow. It maintains a secure execution boundary by separating analysis and preview stages from actual file modifications, ensuring user intent is preserved.
[SAFE]: All dependencies and references are either standard libraries or internal resources associated with the skill author, and no remote code execution or data exfiltration patterns were found.

lov-checkpoint