corp-new

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands using git, gh (GitHub CLI), and rg (ripgrep) to perform repository operations and file indexing. These operations are essential to its primary function and are protected by explicit 'Stop and ask' instructions that require user confirmation before any state changes (create, clone, commit, push, or file edit).
  • [PROMPT_INJECTION]: The skill defines a safety model that explicitly prohibits destructive actions (no force-push, no deletion) and ensures the agent respects user worktrees. It uses clear boundary instructions to prevent the agent from overriding its safety protocols.
  • [DATA_EXFILTRATION]: While the skill interacts with the GitHub API via the gh CLI, it is configured to create private repositories by default and verifies this visibility during the validation phase. No unauthorized data transmission or exfiltration patterns were observed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 04:36 PM
Security Audit — agent-trust-hub — corp-new