lgnh-branch-clear-gone
Pass
Audited by Gen Agent Trust Hub on Jul 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes bash commands including Git operations and shell processing tools like grep, sed, and awk to manage local repository state.
- [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface (Category 8) as it interpolates data from Git command output into shell commands without sanitization. * Ingestion points: Branch names are extracted from the output of
git branch -vandgit worktree listin SKILL.md. * Boundary markers: No boundary markers or ignore-instructions are used for the data processed from command output. * Capability inventory: The skill has the capability to delete directories (git worktree remove) and Git branches (git branch -D). * Sanitization: No validation or escaping is applied to the branch names before they are used as variables in shell commands.
Audit Metadata