The Agent Skills Directory

[SAFE]: The monitoring script operates locally by tracking tool call counts in a session-specific file within the system's temporary directory. This behavior does not involve network access, sensitive file manipulation, or credential harvesting.
[SAFE]: The skill contains references to well-known technology platforms for further reading on token optimization. These are static links for documentation and do not result in the automated download or execution of remote code.
[SAFE]: Although there is a minor discrepancy between the documentation (mentioning a Node.js script) and the provided code (a Bash script), the logic in the script is benign and intended for basic session state tracking.
[SAFE]: The skill does not attempt to override safety guidelines or bypass agent constraints; its primary purpose is to advise the user on utilizing standard built-in commands for context management.

strategic-compact