advisorlisted

# Advisor (GPT-6 Astra) This is the Codex-native advisor maintained in this library. It is a single-turn, independent, read-only consult — not delegated implementation and not a substitute for the lead's judgment. <p align="center"><img src="assets/architecture.svg" alt="Codex advisor: the solver (any Codex lead) sends a self-contained briefing to an Astra xhigh advisor; the advisor returns one decisive read-only review" width="900"></p> Use a new output path or directory for each run; preserve earlier results and existing user edits. ## Default | Advisor | Effort | Use when | |---|---|---| | `gpt-6-astra` | `xhigh` (Extra high) | A substantive interpretation, approach, draft, analysis, or completion check needs an independent review. | Use Astra at `xhigh` for demanding consults. For a routine question, lower the effort or explicitly choose Terra when the review remains adequate. Reserve `max` for a demonstrated need; Astra does not support `none`. A fresh Astra session is a separate review context, not an independent model family when the caller also uses Astra. Luna is not an advisor tier. It is appropriate only for tightly specified mechanical work with objective acceptance checks, not for judgment or synthesis. ## When to consult - Before committing to a substantive interpretation, writing/analysis strategy, or change of approach. - When stuck: repeated errors, a non-converging approach, or results that do not fit. - Before declaring a multi-step task complete,