persistent-worktrees
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and provides architectural guidance for the AI agent to manage local filesystems appropriately. It promotes data persistence and stability by avoiding ephemeral storage like /tmp or ramfs.
- [COMMAND_EXECUTION]: The skill includes benign shell examples using standard Git and filesystem commands (
mkdir,git worktree add) to demonstrate safe practices. These commands are localized to the user's home directory and do not involve privilege escalation. - [DATA_EXPOSURE]: There are no patterns of data exfiltration or sensitive information exposure. The instructions emphasize using read-only tools for filesystem inspection and respect user privacy by suggesting asking the user when storage locations are ambiguous.
- [REMOTE_CODE_EXECUTION]: No external downloads, remote script execution, or package installations are present in the skill files.
Audit Metadata