write-refactor

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines a strict procedural guide for code refactoring, emphasizing safety and verification.
  • [SAFE]: Instructions require the agent to use predefined commands from a local AGENTS.md file and explicitly ask the user if any required command is missing.
  • [SAFE]: The methodology encourages 'Deletion Safety' by requiring the agent to provide evidence of zero callers through grep and string searches before removing code.
  • [SAFE]: There are no instances of prompt injection, data exfiltration, or obfuscation. The skill is entirely focused on software engineering best practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 02:48 PM
Security Audit — agent-trust-hub — write-refactor