skills/jehudarajasa/skills/cut-branch/Gen Agent Trust Hub

cut-branch

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests issue titles and labels from an external tracker to determine the branch name.
  • Ingestion points: Data is read from ../../references/issue-tracker.md or provided via user arguments (ID/URL).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified for the issue content.
  • Capability inventory: The skill executes shell commands (git checkout).
  • Sanitization: The skill provides logic to condense the title into a slug that is lowercased and hyphen-separated, which significantly mitigates command injection risks during the git operation.
  • [COMMAND_EXECUTION]: The skill performs local shell execution to create and switch branches.
  • Evidence: The instruction git checkout -b <branch-name> in SKILL.md directly executes a command using variables derived from external issue tracker data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 10:13 AM