finishing-a-development-branch

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates legitimate developer operations using local command-line tools. Its behavior is consistent with its stated purpose of managing branch lifecycles.
  • [COMMAND_EXECUTION]: The skill utilizes common version control tools (git) and the GitHub CLI (gh). These commands are standard for software development and are executed based on explicit user selection from a structured menu.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a minimal attack surface for indirect injection as it processes user input for workflow selection and confirmation.
  • Ingestion points: User input for option selection (1-4) and the confirmation string 'discard' in SKILL.md.
  • Boundary markers: The skill uses clear, closed-ended prompts and requires a specific literal string ('discard') to confirm destructive operations.
  • Capability inventory: Local file system access via git, branch deletion, and PR creation via the GitHub CLI.
  • Sanitization: Inputs are validated against a strict set of expected values (1, 2, 3, 4, or 'discard').
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 05:04 AM
Security Audit — agent-trust-hub — finishing-a-development-branch