正在打开每日关注
正在打开每日关注
一条可复制的 Codex 配置提示词：分层模型树加按需唤起的架构师代理，降低编码成本。
Codex tip: once GPT-6.1 Sol is your main model, stop running Astra on every turn put Astra on call as an architect agent GPT-6.1 Sol keeps writing the code Astra only gets spawned at three points: → before a plan: is this the right approach? → when the same error comes back: am I digging in the wrong place? → before "done": what did I miss? Astra reviews. Sol ships Jev engineering is the same move one layer down: the forks that need no thinker (which file, which tool, retry or stop) go to Jev in under half a second, and the big models only see the ones that split - the full tree > GPT-6.1 Sol on high runs the main session > explorer reads the code on Luna > worker edits and runs tests on Sol > researcher pulls the docs on Luna > all three on medium > Astra on call as the architect > auto_review checks every approval paste the tree and this prompt into Codex ↓ "Rebuild my Codex setup around this tree: 1. Check ~/.codex/agents and .codex/agents for agents that already fit explorer, worker and researcher. > Draft new TOML files only for missing roles > explorer and researcher on gpt-6-luna, worker on gpt-6.1-sol, all with model_reasoning_effort medium > Add an architect agent on gpt-6-astra, model_reasoning_effort high, whose only job is reviewing plans, repeated errors and finished work > Skip any that pin a different model and list them 2. In ~/.codex/config.toml set model to gpt-6.1-sol, model_reasoning_effort to high and approvals_reviewer to auto_review 3. Find anything that would override this (active profiles, flags in my shell aliases, agents.default_subagent_model). Report it, change nothing 4. Add one rule to AGENTS.md: spawn the architect before a large plan, when an error repeats, and before calling a long task done Show me every change as a diff first. No edits until I say go." ↳ https://t.co/eyMF8gZKLE
是什么：一条可复制的 Codex 多智能体配置提示词与配套架构图。作者主张用 GPT-6.1 Sol 以高推理强度跑主会话负责规划与写码，explorer/worker/researcher 以 medium 强度并行分工（读码、改码跑测试、拉文档），Astra 作为「架构师」按需唤起，auto_review 审查每次批准，并用一段结构化 Prompt 让 Codex 自检现有 agents、补齐缺失角色的 TOML、改写 config.toml 并在 AGENTS.md 加入触发规则。
核心设计：三个机制——(1) 角色分层：贵模型只做规划写码，读码/查文档降级到低成本模型；(2) 事件触发的审查：架构师只在「定计划前、错误重复时、宣告完成前」介入，而非每轮陪跑，且「永不亲自写码，由 Sol 落实其意见」；(3) Jev 分流层：把「改哪个文件、用哪个工具、重试还是停」这类高置信决策（p≈0.87–0.97）直接在代码层解决，模糊的才 split 给主模型。三帧图显示 forks 从 1,365 涨到 1,705、Astra 调用与 token 同步上升，呈现运行时可观测状态。
关键洞察：「Astra reviews, Sol ships」——把审查做成门控事件而非持续运行，是质量与成本兼顾的编排思路；「先看 diff，未确认不改动」体现了对代理修改本机配置的安全边界。
边界与风险：GPT-6.1 Sol/Luna/Astra 等模型名无法在公开资料中核实，Jev 概率数据属演示性质；引用推文与 X 文章不可访问，外链 ChatGPT Learn 页面基本是导航目录，佐证有限；照搬该 Prompt 会改动 ~/.codex 配置，务必人工审阅 diff 并留意作者提醒的 profiles、shell 别名等覆盖项。