# Practical multi-agent orchestration in Codex
**作者**: eric provencher
**日期**: 2026-07-24T17:30:50.000Z
**来源**: [https://x.com/pvncher/status/2080707291603407077](https://x.com/pvncher/status/2080707291603407077)
---
GPT-5.6 Sol gets especially interesting when it has a team to work with. Codex's new Multi-Agent V2 tools give Sol and Terra a natural way to delegate tasks, share updates, and coordinate through complex tasks.
Ultra makes agent coordination the default and is best reserved for high-stakes work, where ambiguity or scattered context justify the added depth.
For other tasks, a short prompt or skill can encourage the same collaborative behavior from Sol Medium, as it stays in conversation with you while organizing the work behind the scenes. With the right nudge, Sol can turn broad requests into focused assignments, bringing in other agents, and deciding when a problem calls for deeper reasoning.
## Match reasoning to the work
While it's possible to have Sol delegate to another model like Terra, the simplest setup keeps one model family and adjusts only the reasoning effort, with dedicated roles like so:
- Scout — GPT-5.6 Sol Light. Answer narrow, read-only questions: locate files, trace a code path, or find relevant tests.
- Worker — GPT-5.6 Sol Medium. Implement scoped changes, run checks, or handle supporting work.
- Smart worker — GPT-5.6 Sol High. Take on difficult implementation, resolve ambiguity, or coordinate help when useful.
Treat these roles as useful defaults. Sol Light retains the judgment to find useful context without spending as much reasoning on discovery.
## Let the team coordinate
The coordinator acts as the primary delegator: it assigns substantive work, avoids duplicate investigations, and tracks what each agent is doing. Scouts can investigate in parallel, while workers can share implementation when responsibilities are clear.
Agents can also message one another directly through a common messaging system with separate inboxes. When a scout discovers something a worker needs, it can recognize the dependency and pass along its findings without waiting for the coordinator to relay them.
Concurrency is configurable per thread and defaults to four agents, including the coordinator. Within that budget, a smart worker might coordinate a scout and another worker, or the coordinator could send three scouts to investigate separate questions.
## Choose what context agents inherit
Forking conversation history helps agents understand the broader goal and earlier decisions. Starting with fork_turns: "none" gives an agent a fresh, focused assignment instead. Fresh-context agents can still recognize when a teammate needs information and contact them independently.
Agents that inherit their parent's context may also see its orchestration instructions. When an agent should remain a leaf, give it a short boundary:
> Complete this assignment directly. Do not spawn other agents; your parent's delegation instructions apply only to your parent.
Fresh-context agents won't inherit task-specific tool or safety boundaries, so include any essential restrictions directly in their assignments.
## Capture the pattern in a skill
A practical skill can give the coordinator a few standing instructions:
> Stay available to the user while delegating substantive work. Send focused, read-only scouts out in parallel with reasoning_effort: "low" and fork_turns: "none". Use reasoning_effort: "medium" for routine implementation and reasoning_effort: "high" for harder problems. Give each agent clear ownership, avoid overlapping assignments, and tell leaf workers not to delegate. Bring the results together and keep approvals with the user.
## Experiment with the knobs
Start with these defaults, then experiment with reasoning effort, context inheritance, delegation authority, and how agents collaborate. The goal is to understand which settings help a team move the work forward without spending more reasoning than the task requires.
## 相关链接
- [eric provencher](https://x.com/pvncher)
- [@pvncher](https://x.com/pvncher)
- [59K](https://x.com/pvncher/status/2080707291603407077/analytics)
- [Upgrade to Premium](https://x.com/i/premium_sign_up)
- [1:30 AM · Jul 25, 2026](https://x.com/pvncher/status/2080707291603407077)
- [59.4K Views](https://x.com/pvncher/status/2080707291603407077/analytics)
- [View quotes](https://x.com/pvncher/status/2080707291603407077/quotes)
---
*导出时间: 2026/7/25 11:33:47*
---
## 中文翻译
# Codex 中的实用多智能体编排
**作者**: eric provencher
**日期**: 2026-07-24T17:30:50.000Z
**来源**: [https://x.com/pvncher/status/2080707291603407077](https://x.com/pvncher/status/2080707291603407077)
---
当 GPT-5.6 Sol 拥有一个团队协作时，情况变得尤为有趣。Codex 全新的多智能体 V2 工具为 Sol 和 Terra 提供了一种自然的方式来委派任务、共享更新，并通过复杂任务进行协调。
Ultra 将智能体协调设为默认模式，最适合用于高风险工作，因为在这些场景下，模糊性或分散的上下文证明了增加深度的合理性。
对于其他任务，简短的提示或技能即可鼓励 Sol Medium 表现出同样的协作行为，它在与你保持对话的同时，在后台组织工作。只需适当的引导，Sol 便可以将宽泛的请求转化为集中的任务，引入其他智能体，并判断何时需要更深层的推理来解决某个问题。
## 将推理与工作相匹配
虽然可以让 Sol 委派给 Terra 这样的其他模型，但最简单的设置是保持单一模型系列，仅调整推理投入，并设定如下特定角色：
- Scout（侦查员）— GPT-5.6 Sol Light。回答狭窄的只读问题：定位文件、追踪代码路径或查找相关测试。
- Worker（工作者）— GPT-5.6 Sol Medium。实施特定范围的更改、运行检查或处理辅助工作。
- Smart worker（智能工作者）— GPT-5.6 Sol High。承担困难的实施工作、解决模糊性问题，或在必要时协调寻求帮助。
可以将这些角色视为有用的默认选项。Sol Light 保留了发现有用上下文的判断力，而无需在发现环节消耗过多的推理。
## 让团队进行协调
协调员作为主要的委派者：它分配实质性工作、避免重复调查，并跟踪每个智能体的任务。侦查员可以并行调查，而工作者可以在职责明确时共享实施工作。
智能体还可以通过一个带有独立收件箱的通用消息系统直接相互发送消息。当侦查员发现工作者需要的信息时，它可以识别这种依赖关系，并直接传递其发现，而无需等待协调员进行中转。
并发性是按线程可配置的，默认为包括协调员在内的四个智能体。在该预算范围内，一个智能工作者可以协调一个侦查员和另一个工作者，或者协调员可以派出三个侦查员分别调查不同的问题。
## 选择智能体继承的上下文
分叉对话历史有助于智能体理解更广泛的目标和先前的决策。以 fork_turns: "none" 开始，则是给智能体一个全新、集中的任务。全新上下文的智能体仍然可以识别队友何时需要信息，并独立与他们联系。
继承其父级上下文的智能体可能也会看到其编排指令。当一个智能体应保持为叶子节点时，请给它一个简短的界限：
> 直接完成此任务。不要生成其他智能体；你父级的委派指令仅适用于你的父级。
全新上下文的智能体不会继承任务特定的工具或安全边界，因此请在他们的任务中直接包含任何必要的限制。
## 在技能中捕获这种模式
一个实用的技能可以给协调员一些常驻指令：
> 在委派实质性工作时保持对用户的可用性。并行派出专注的只读侦查员，并设置 reasoning_effort: "low" 和 fork_turns: "none"。对于常规实施使用 reasoning_effort: "medium"，对于更困难的问题使用 reasoning_effort: "high"。给每个智能体明确的所有权，避免重叠的任务，并告知叶子工作者不要进行委派。汇总结果并将审批权留给用户。
## 尝试调整各种参数
从这些默认设置开始，然后尝试调整推理投入、上下文继承、委派权限以及智能体的协作方式。目标是理解哪些设置可以帮助团队推进工作，而无需消耗超出任务所需的推理资源。
## 相关链接
- [eric provencher](https://x.com/pvncher)
- [@pvncher](https://x.com/pvncher)
- [59K](https://x.com/pvncher/status/2080707291603407077/analytics)
- [Upgrade to Premium](https://x.com/i/premium_sign_up)
- [1:30 AM · Jul 25, 2026](https://x.com/pvncher/status/2080707291603407077)
- [59.4K Views](https://x.com/pvncher/status/2080707291603407077/analytics)
- [View quotes](https://x.com/pvncher/status/2080707291603407077/quotes)
---
*导出时间: 2026/7/25 11:33:47*