The Agent Skills Directory

[SAFE]: The skill's primary function is to audit the agent's internal memory and local project configuration files (MEMORY.md, CLAUDE.md, and .claude/rules/). It uses standard, non-malicious shell commands like pwd, ls, and find to locate and inspect these files.
[DATA_EXPOSURE]: While the skill reads sensitive internal memory files, this is done for the explicit purpose of reporting back to the user about their own project's history. No data is sent to external servers or domains.
[COMMAND_EXECUTION]: The shell commands used for file path resolution (sed) and checking for file existence (find) are used safely to facilitate the memory review process and do not include any user-controlled input that could lead to command injection.

memory-review