Skills
skills/ag-grid/ag-charts/git-worktree-clean/Gen Agent Trust Hub

git-worktree-clean

Warn

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill uses string interpolation to build a shell command (git reset --hard "$TARGET_BRANCH") using the ARGUMENTS variable provided by the user. This approach is susceptible to command injection, as shell metacharacters such as backticks or $() can be used to execute arbitrary code within the shell context.
  • [INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface for indirect prompt injection by processing untrusted user input that influences command execution.
  • Ingestion points: ARGUMENTS parameter in SKILL.md.
  • Boundary markers: None.
  • Capability inventory: Shell command execution via Git CLI.
  • Sanitization: No input validation or character filtering is implemented.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 12, 2026, 05:42 PM