The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes standard git commands including git status, git add, git diff, and git commit. It also dynamically detects and runs project-specific verification commands (such as linting or testing) found in configuration files like package.json or Makefile. These actions are core to the skill's documented workflow.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the ingestion and analysis of untrusted data from the local repository environment.
Ingestion points: It reads content from package.json, Makefile, and the output of git diff --staged to generate commit messages and suggest logical splits.
Boundary markers: No explicit delimiters or instructions are provided to the agent to prevent it from following malicious instructions that might be embedded within project files or code diffs.
Capability inventory: The skill can execute shell commands and modify the repository's git state.
Sanitization: The skill does not perform validation or sanitization of the content read from the repository before processing it.

commit-elegant