The Agent Skills Directory

[SAFE]: The skill implements a 'deny-list' approach in scripts/bundle.py and scripts/bundle_zip.py to prevent the accidental inclusion of sensitive files such as .env and __pycache__ in generated bundles.
[SAFE]: Command execution is handled securely; the skill uses the Python zipfile library for archive creation rather than spawning shell processes (e.g., zip or tar), which mitigates risk of command injection.
[SAFE]: File path resolution in scripts/path_resolver.py uses heuristic landmark checks (like .git or .agent) to define the project root, preventing unauthorized directory traversal outside of the intended workspace.

zip-bundling