The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The skill uses Bash to execute standard Git operations (fetch, branch). These actions are confined to the intended purpose of repository maintenance and follow the provided instructions.
[INDIRECT_PROMPT_INJECTION] (LOW): The skill ingests data (branch names) from potentially untrusted remote sources. While no explicit boundary markers or sanitization logic are present in the provided command, Git's internal validation of branch names—which restricts the use of control characters, spaces, and shell-sensitive symbols like semicolons or pipes—serves as a natural and effective mitigation against command injection in this context.
[DATA_EXFILTRATION] (SAFE): No patterns of unauthorized data access, network exfiltration, or hardcoded credential exposure were identified.

git-prune