The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to execute the git commit command to finalize changes. This action is a core functionality of the skill and is strictly gated by a mandatory user confirmation step in the interaction flow.
[PROMPT_INJECTION]: The skill reads output from git diff, creating a surface for indirect prompt injection if the code changes contain malicious instructions in comments or strings. However, the risk is mitigated by the fact that the agent's output is reviewed by the user before execution. 1. Ingestion points: staged and working tree diffs (SKILL.md). 2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present. 3. Capability inventory: Command execution via git commit (SKILL.md). 4. Sanitization: No specific sanitization or filtering of the diff content is implemented.

git-commit-doc