The Agent Skills Directory

[SAFE]: The skill is a standard administrative utility for managing a local RLM (Resource Ledger Management) system. Access is limited to local cache and manifest files for the purpose of synchronization. There are no hardcoded credentials or network operations to external domains.\n- [COMMAND_EXECUTION]: The skill executes local Python scripts (cleanup_cache.py, inventory.py) within the vendor's plugin directory structure. The execution flow is protected by a mandatory dry-run step and requires explicit user consent before applying changes.\n- [PROMPT_INJECTION]: Instructions are focused on operational safety and role-adherence, with no attempts to override agent safety filters or extract system instructions.

rlm-cleanup