The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local shell commands including git diff, git status, and git commit to inspect code changes and manage the repository state.
[INDIRECT_PROMPT_INJECTION]: The skill is exposed to indirect prompt injection because it processes untrusted file content during the diff analysis phase.
Ingestion points: Staged changes are read into the agent context using git diff --cached in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to prevent the agent from obeying commands embedded within the code diffs.
Capability inventory: The skill has the capability to execute git commands and create commits, which could be exploited if the agent is tricked by malicious diff content.
Sanitization: There is no evidence of sanitization or filtering applied to the diff output before it is analyzed by the language model.

git-commit-pro