refactor

Pass

Audited by Gen Agent Trust Hub on Jul 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands to integrate with the development environment, running test suites (e.g., bun test, pytest, go test), type checkers (e.g., tsc --noEmit), and linters (e.g., eslint) to verify code changes after each refactoring step.- [COMMAND_EXECUTION]: It utilizes a local script scripts/ast_grep_helper.py and the sg CLI for structural code transformations, executing these tools via the shell to perform AST-based replacements.- [DATA_EXFILTRATION]: The skill manages orchestration by interacting with configuration files in the user's home directory (e.g., ~/.omo/teams/refactor-squad/config.json). This is used for coordinating sub-agents in team mode.- [PROMPT_INJECTION]: The skill's architecture for analyzing and refactoring codebases presents an indirect prompt injection surface. Instructions or patterns embedded in the analyzed code could potentially influence the refactoring plan, although the skill includes multiple manual and automated verification steps to mitigate this risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 19, 2026, 05:10 AM
Security Audit — agent-trust-hub — refactor