The Agent Skills Directory

[SAFE]: The skill uses yaml.safe_load() in its indexing script (scripts/index.py) to process metadata, which is a secure practice that mitigates risks associated with untrusted YAML deserialization.
[COMMAND_EXECUTION]: A subprocess.run call exists within tests/test_compression.py. This call is used exclusively for automated testing to execute the local index.py script and does not expose the system to arbitrary command injection from external input.
[DATA_EXFILTRATION]: All data is stored locally in the ./memory/ directory. There are no network requests, remote downloads, or features that would allow data to be transmitted to an external server.

memory-bank