The Agent Skills Directory

[SAFE]: The skill uses atomic file operations (temporary file write followed by rename) and implements a rollback mechanism in scripts/file_operations.py to ensure repository integrity during multi-file updates.
[SAFE]: Strict path validation is enforced in scripts/file_operations.py and scripts/governance_planning.py, preventing directory traversal and symlink-based escape attacks by verifying all targets are within the repository root.
[COMMAND_EXECUTION]: Local shell command execution is limited to the test suite and validation scripts (scripts/test_governance_support.py). These calls are implemented securely using argument lists rather than shell strings, avoiding command injection vulnerabilities.
[SAFE]: The package includes a defensive checker (scripts/check.py) designed to identify and flag host-specific paths or environmental markers within the skill's own source, minimizing the risk of accidental information disclosure.
[SAFE]: All logic is implemented using standard Python libraries. The skill does not download external packages, execute remote code, or perform network operations.

repo-governance