The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (scripts/lint_wiki.py, scripts/append_log.py, scripts/secret_scan.py, and scripts/session_text.py) and basic shell builtins via Bash commands to discover the wiki workspace, format page templates, maintain audit histories, and parse existing session data.
[SAFE]: Secret exposures are mitigated dynamically by passing staged outputs into a regex-based token validator (scripts/secret_scan.py) before write workflows commit them to the immutable raw data store. No remote network connections or external third-party tools are utilized during its normal operation flow.

llm-wiki