Empire LLM for Codex

Empire LLM gives Codex bounded external-model reviews, quarantined proposals, benchmark research charts, cost-approved media generation, secure local settings, and paid-response recovery. Public-web research uses Codex native tools only and stops at access restrictions; third-party web extraction and managed browser actions are removed in 1.7.2. Artificial Analysis Free supports unverified research charts; exact benchmark-to-OpenRouter qualification requires explicit Pro/Commercial identity evidence. Codex retains repository and decision authority. Users supply model-provider credentials; external model and media usage may incur charges.