The Agent Skills Directory

[SAFE]: No security issues or malicious patterns were detected during the analysis of the skill.
[COMMAND_EXECUTION]: The skill uses the git and gh command-line utilities for version control and pull request management. These commands are used according to their intended design for repository automation and are not used to perform unauthorized actions.
[PROMPT_INJECTION]: The skill analyzes untrusted data from git diff to generate commit messages and pull request bodies. While this represents an indirect prompt injection surface, the risk is minimal in this context as the output is used for text generation within a standard developer workflow. (1) Ingestion points: git diff output. (2) Boundary markers: None. (3) Capability inventory: git commit, git push, gh pr create. (4) Sanitization: None.

github-pr