The Agent Skills Directory

[SAFE]: The skill manages internal state for skill loadouts by writing to a dedicated directory (~/.openclaw/skill-state/skill-loadout-manager/). Access to the user's home directory is limited to this specific path for state management.
[SAFE]: The Python script uses yaml.safe_load for all data ingestion from configuration files, which is a security best practice to prevent arbitrary object instantiation.
[SAFE]: The skill does not perform any network operations, external downloads, or arbitrary command execution. All operations are local file system reads and writes related to its stated purpose.
[SAFE]: The skill includes discovery logic for other skills but restricts its analysis to reading the first few lines of SKILL.md files to extract metadata, posing no risk of executing untrusted content.
[SAFE]: No evidence of prompt injection, obfuscation, or persistence mechanisms (such as crontab or shell profile modification) was found in the provided files.

skill-loadout-manager