The Agent Skills Directory

Indirect Prompt Injection (HIGH): The skill has a high-risk attack surface for indirect injection. It processes untrusted data from the local repository and has the authority to execute state-changing git commands.
Ingestion points: The skill reads file contents and changes using git status and git diff.
Boundary markers: There are no instructions provided to the agent to treat file content as data only or to ignore instructions found within the code diffs.
Capability inventory: The skill can execute git add and git commit to modify the repository state.
Sanitization: No sanitization or validation of the file content is performed before the agent processes it to generate commit messages or decide on staging.
Command Execution (LOW): The skill uses standard git commands (git status, git diff, git add, git commit). While these are legitimate for a git-focused skill, they provide the mechanism through which an indirect prompt injection could be weaponized.

commit