Set these in the config.toml for codex and you'll get a lot more info while it's running: model_reasoning_summary = "detailed" model_verbosity = "high"

Set these in the config.toml for codex and you'll get a lot more info while it's running: model_reasoning_summary = "detailed" model_verbosity = "high" model_supports_reasoning_summaries = true show_raw_agent_reasoning = true