The Agent Skills Directory

[SAFE]: The skill provides purely instructional content for Git version control management, including branching strategies and commit conventions.
[SAFE]: Includes security best practices by recommending that sensitive files like .env be added to the .gitignore file to prevent accidental credential exposure.
[SAFE]: Promotes safe Git operations by recommending git push --force-with-lease instead of the more destructive --force flag.
[INDIRECT_PROMPT_INJECTION]: The skill documents workflows for handling data from external sources (such as pull request descriptions and commit messages). While this represents a theoretical attack surface for indirect prompt injection if an agent were to autonomously act on that data, the skill itself does not include any automated execution or unsafe data processing scripts.
Ingestion points: PR descriptions and commit messages (SKILL.md).
Boundary markers: Not explicitly defined for these data types.
Capability inventory: Standard Git command-line operations.
Sanitization: None specified; relies on user review of PR content.

git-workflow