The Agent Skills Directory

[SAFE]: The skill implements protective measures to prevent accidental data loss during git operations by enforcing a workflow that requires 'git status' and 'git diff' before any potentially destructive command.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local git repository metadata (filenames, commit messages, and diffs) which could potentially contain adversarial instructions. However, the instructions mandate that the agent performs manual file edits and requests explicit user confirmation for specific commands, which serves as a robust control against automated exploitation of injected content.

git-safety