The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes several local git commands (git status, git diff --cached, git diff, and git log) to analyze the repository's current state and history. These commands are legitimate and necessary for the skill's intended purpose.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes content that could be influenced by external actors (e.g., in a collaborative repository). * Ingestion points: Untrusted data enters the context through the output of git diff and git log commands. * Boundary markers: The skill does not use specific delimiters or instructions to prevent the agent from following commands that might be embedded within the source code diffs or commit history. * Capability inventory: The skill can execute shell commands (git) and propose commit messages for execution. * Sanitization: There is no sanitization or escaping of the content fetched from the git repository before it is provided to the agent for analysis.

git-commit