tidy-up
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is designed to execute potentially destructive shell commands to clean up the local filesystem.
- Evidence: The SKILL.md file and references/targets.md define commands that utilize
rm -rf,git clean -fdX,git branch -D, andgit worktree prune. While these are the primary purpose of the skill, they carry inherent risk if misused by an autonomous agent. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted filesystem data and possesses high-privilege capabilities, creating a potential surface for indirect prompt injection.
- Ingestion points: Untrusted filenames, directory paths, and git branch names are ingested via discovery commands (e.g.,
git worktree list,git branch -vv) as described in references/targets.md. - Boundary markers: Present. The skill mandates an inventory table with risk labels and user confirmation before deletion as specified in the Default Flow and safety.md.
- Capability inventory: High-privilege shell commands including
rm -rf,git clean -fdX, andgit branch -Dare used for cleanup tasks. - Sanitization: Absent. No automated filtering or sanitization of filesystem strings is documented; the skill relies on the human-in-the-loop verification of the inventory table.
Audit Metadata