The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/git_stats.py executes the git binary using the subprocess.run method.
The implementation uses a list of arguments and defaults to shell=False, which is a secure way to prevent shell injection attacks.
User-supplied inputs (like dates or authors) are interpolated into specific Git flags (e.g., --since={value}), which prevents them from being interpreted as separate command-line arguments.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a surface for indirect prompt injection because it ingests and processes data from external Git repositories.
Ingestion points: The tool reads commit messages, contributor names, and file paths from the repository being analyzed.
Boundary markers: The output is printed as plain text without specialized delimiters or instructions to the AI agent to ignore potentially malicious content embedded in the git logs.
Capability inventory: The skill allows the agent to execute Git commands and read the full history of a repository.
Sanitization: The script performs no sanitization or filtering of commit messages or author metadata before returning the data to the agent's context.

git-stats