finishing-a-development-branch

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute standard version control operations (git), GitHub CLI commands (gh), and various language-specific test runners (npm, cargo, pytest, go). These actions are limited to the intended scope of branch lifecycle management.
  • [PROMPT_INJECTION]: Analysis of the instructions and metadata revealed no attempts to bypass safety guardrails or hijack agent behavior.
  • [DATA_EXFILTRATION]: No patterns of sensitive data access or exfiltration to unauthorized remote servers were detected. The skill only interacts with local project data and the configured Git origin.
  • [SAFE]: The skill implements safety-first practices, such as mandatory test checks before merging and explicit user confirmation for destructive actions like deleting branches.
  • [SAFE]: Indirect Prompt Injection Assessment: 1. Ingestion points: Git branch names and test runner outputs (SKILL.md). 2. Boundary markers: Absent for external tool outputs. 3. Capability inventory: Branch deletion and PR creation via Bash. 4. Sanitization: No sanitization is performed on test output or metadata. Risk is low as the agent acts within the confines of standard Git operations.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 02:45 PM