lf-git-checkout

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses user-provided $ARGUMENTS and local branch names directly in shell commands such as git checkout <BRANCH_DESTINO>. If these inputs contain shell metacharacters (e.g., semicolons, backticks), they could lead to unintended command execution in the host environment.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through external data processed at runtime.
  • Ingestion points: User-supplied input via $ARGUMENTS and repository metadata retrieved via git branch and git submodule status in SKILL.md.
  • Boundary markers: None; external content is interpolated directly into commands and presented to the agent without delimiters or safety warnings.
  • Capability inventory: The skill executes various git and bash commands, including file system navigation, stashing, and repository state modification across multiple directories (SKILL.md).
  • Sanitization: No input validation or shell escaping is performed on the data before it is used to construct executable commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 12:15 AM