The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill operates on untrusted data from the repository's current state.
Ingestion points: The agent processes output from git diff --staged and git status to generate commit messages (SKILL.md).
Boundary markers: The instructions do not define clear boundaries or provide warnings to the agent to disregard instructions embedded within the code changes being reviewed.
Capability inventory: The skill possesses the capability to execute shell commands, specifically git operations that can modify the repository state via commits (SKILL.md).
Sanitization: The implementation employs a quoted heredoc ('EOF') when passing the message to the git commit command, which effectively prevents shell command injection from the generated commit message content.

commit