The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/gitlab_url_guard.py invokes the git CLI to determine the repository root and retrieve remote URLs. These calls are performed using subprocess.run with a fixed list of arguments, avoiding shell execution and protecting against command injection.\n- [DATA_EXFILTRATION]: The skill processes project-specific configuration stored in gitlab-url-guard.json and performs text normalization on user-provided files. No unauthorized network activity or access to sensitive system credentials was detected.\n- [PROMPT_INJECTION]: The skill has a data ingestion surface where it reads external text files for validation and normalization. While this allows for potential indirect prompt injection, the script does not execute or interpret the content as instructions; it only applies regex-based link replacement.\n- [REMOTE_CODE_EXECUTION]: No patterns of remote script downloading or dynamic code execution (such as eval or exec) were found in the provided files.

gitlab-url-guard