Getting the most out of Opus 5.5 in Claude and Claude Code

Opus 5.5 works well with the way you already use Claude. A few things behave differently, though: it works for longer on its own, it tells you plainly what it did, and it thinks before every reply. This guide covers how to work with Opus 5.5 in Claude apps and Claude Code, including how to prompt the model, steer a long run, and check your results.Opus 5.5 能够很好地适配你现有的 Claude 使用习惯。不过，有几点表现有所不同：它能更长时间地独立工作，会明确告诉你它做了什么，并且在每次回复前都会进行思考。本指南涵盖了如何在 Claude 应用和 Claude Code 中使用 Opus 5.5，包括如何向模型下达指令、引导长任务运行以及检查结果。

TRY THIS FIRST

Three things to try in your first session with Opus 5.5首次使用 Opus 5.5 时可以尝试的三件事

1. Hand over the whole task. Say what “done” looks like and when you want it to stop and ask. Then let it work.将整个任务交给它。说明“完成”的标准是什么，以及你希望它在何时停止并询问你。然后，让它开始工作。
2. Delete “think carefully” lines. Opus 5.5 already thinks before every reply.删除“仔细思考”之类的语句。Opus 5.5 在每次回复前已经会自动进行思考。
3. When a long run ends, read what it needs from you first.当长任务结束时，先阅读它需要你处理的内容。

1. HOW TO ASK1. 如何提问

Say what “done” looks like, then let it run明确“完成”的标准，然后让它运行

What to do. Give the whole task in one message. Name the finish line, like “the tests pass” or “every endpoint is migrated.” Then let it cook.怎么做：在一条消息中交代完整任务。设定终点线，例如“测试通过”或“所有端点已迁移”。然后让它去执行。

Why it matters on Opus 5.5. Opus 5.5 keeps going on long, multi-part work better than Opus 5 did. Compared to prior Opus models, its biggest gains are on multi-step work, like carrying a change through a large repository until the tests pass. Early testers had it run long coding tasks for hours with little oversight. With a clear finish line, it knows when it’s done.为什么这对 Opus 5.5 很重要：与 Opus 5 相比，Opus 5.5 在处理长周期、多阶段任务时表现更稳健。相较于以往的 Opus 模型，它在多步骤任务上的提升最为显著，例如在一个大型代码库中持续进行修改直到测试通过。早期测试人员让它在极少人工干预的情况下运行了数小时的编程任务。只要有清晰的终点线，它就知道何时该结束。

How. In Claude Code, for example:如何操作：以 Claude Code 为例：

Migrate the payment endpoints from the old client to the new one.
Done means: every endpoint uses the new client, the old client is deleted, and the test suite passes.
Stop and ask me only if a test fails for a reason you can't explain.

Stop telling it to “think hard”停止要求它“深度思考”

What to do. Remove “think carefully,” “think step by step,” and similar lines from your prompts and your saved instructions.怎么做：从你的提示词和保存的指令中移除“仔细思考”、“一步步思考”及类似的语句。

Why it matters on Opus 5.5. Opus 5.5 always thinks before it replies, and it decides how much. You don’t need to ask it to think. In our testing in a chat product, removing a “think carefully” line made replies start sooner, with no clear drop in quality.为什么这对 Opus 5.5 很重要：Opus 5.5 在回复前总是会进行思考，并且它会自行决定思考的深度。你不需要要求它思考。在我们聊天产品的测试中，移除“仔细思考”这一行指令后，回复启动得更快，且质量没有明显下降。

How. Delete the line. For a quick answer to a simple question, say so: “Answer directly.” To change how much it thinks in Claude Code, change effort.如何操作：删除该语句。如果需要对简单问题进行快速回答，可以直接说：“直接回答”。若要在 Claude Code 中调整其思考深度，请更改 effort 参数。

Add to a running task向正在运行的任务中添加要求

What to do. If you remember something mid-run, you can type a follow-up while it works.怎么做：如果你在任务运行中途想起什么，可以在它工作时输入后续指令。

Why it matters on Opus 5.5. Runs are longer now, so a restart costs more.为什么这对 Opus 5.5 很重要：现在的运行周期更长，因此重启任务的代价更高。

How to do it. In Claude Code, type the message and press Enter while Claude works, for example, “Also keep the old endpoint names as aliases.”如何操作：在 Claude Code 中，当 Claude 正在工作时输入消息并按回车键，例如：“同时将旧的端点名称保留为别名。”

For design work, name the styles you don’t want进行设计工作时，列出你不需要的样式

What to do. When you ask for a page, an app, or an artifact, list the design habits you want left out.怎么做：当你要求它制作页面、应用或工件（Artifact）时，列出你希望排除的设计习惯。

Why it matters on Opus 5.5. With no design direction, Opus 5.5 falls back on a few default styles. A general instruction like “avoid a generic look” mostly swaps one default for another. A list of specific patterns works much better.为什么这对 Opus 5.5 很重要：如果没有设计方向，Opus 5.5 会默认采用几种常用风格。像“避免通用外观”这样笼统的指令通常只是把一种默认风格换成另一种。列出具体的模式效果要好得多。

How. Name the patterns:如何操作：列出这些模式：

Build a personal website with placeholder content.
Don't use a cream or off-white background, italic accent words in headings, numbered "01 / 02 / 03" section labels, monospace labels, or pill-shaped buttons.

Then look at what it chose instead. If you don’t like that either, add it to the list and ask again.然后查看它选择的替代方案。如果你依然不满意，将其加入列表并再次提问。

2. STEERING A LONG RUN IN CLAUDE CODE2. 在 CLAUDE CODE 中引导长任务

Tell it which stops you want告诉它你希望在哪些节点停下

What to do. Put a short rule in your CLAUDE.md file about when to stop and ask, and when to keep going.怎么做：在你的 CLAUDE.md 文件中加入一条简短的规则，说明何时该停止并询问，以及何时该继续。

Why it matters on Opus 5.5. Opus 5.5 keeps you posted as it works. On a long task, it sometimes stops to report instead of going on: a summary that names the next step without taking it, an offer to continue, or a list of choices that don’t block the work. It follows instructions that name these stops. Name the stops you want, too.为什么这对 Opus 5.5 很重要：Opus 5.5 会在工作时向你通报进展。在长任务中，它有时会停下来汇报而不是直接继续：比如提供一个总结并指出下一步但不执行，询问是否继续，或者列出不阻塞后续工作的选项。它会遵循这些关于停止节点的指令。你也应该明确你想要的停止点。

How. Add this to CLAUDE.md, and edit it to fit your project:如何操作：将此内容添加到 CLAUDE.md，并根据你的项目进行编辑：

When a step doesn't need my input, keep going. Put status notes in the same message as your next action.
Stop and ask only when you can't continue without me, or before anything destructive: deleting data, force-pushing, or changing anything outside this repository.

If a run stops with “Want me to continue?” reply “continue.” If that happens often, the rule above will help.如果任务因“需要我继续吗？”而停止，回复“continue”。如果这种情况频繁发生，上述规则将会有所帮助。

A rule to keep going means fewer stops, so keep your own check before anything risky or hard to undo. The last line of the rule above does that. Keep permission prompts on for destructive commands too.设置“继续”的规则意味着更少的停顿，因此在执行任何风险较高或难以撤销的操作前，请务必亲自检查。上述规则的最后一行就是为此设置的。对于破坏性命令，请务必保留权限确认提示。

For pair programming, you may want the opposite: a one-line plan before it starts and a short recap at the end. Say that in your CLAUDE.md instead. Opus 5.5 follows either one.对于结对编程，你可能需要相反的设置：在开始前提供一行计划，并在最后进行简短回顾。请在 CLAUDE.md 中注明。Opus 5.5 可以遵循这两种模式。

Ask it to split big work across subagents要求它将大型任务拆分给子智能体（Subagents）

What to do. For an audit, a migration, or a review across a large codebase, ask Opus 5.5 to split the work across subagents and check each result.怎么做：对于大型代码库的审计、迁移或审查，要求 Opus 5.5 将工作拆分给子智能体，并检查每个结果。

Why it matters on Opus 5.5. Early testers had Opus 5.5 coordinate parallel subagents on long audits and migrations, with little oversight.为什么这对 Opus 5.5 很重要：早期测试人员让 Opus 5.5 在几乎无需人工监督的情况下，协调并行子智能体完成了长周期的审计和迁移任务。

How.如何操作：

Audit every service in services/ for the retry bug in the linked issue.
Give each service to its own subagent. When a subagent reports back, check its evidence before you accept it.
Finish with one table: service, affected yes or no, and the evidence.

Keep the task list in a file将任务列表保存在文件中

What to do. For a run that will take a while, ask Opus 5.5 to keep its task list in a file and update it as it goes. Then read the file, not the scrollback, to see where the run is.怎么做：对于耗时较长的任务，要求 Opus 5.5 将任务列表保存在文件中，并随着进度更新。然后通过查看该文件（而不是滚动屏幕）来了解任务进度。

Why it matters on Opus 5.5. Runs are longer now. A long run fills the context window, and Claude Code then summarizes older turns. A list in a file survives that, and it shows you at a glance what’s done and what’s left.为什么这对 Opus 5.5 很重要：现在的运行周期更长。长任务会填满上下文窗口，导致 Claude Code 随后会总结之前的轮次。保存在文件中的列表可以规避这一点，让你一眼就能看出哪些已完成，哪些尚未完成。

How. “Keep a checklist in TASKS.md. Tick each item when it’s done, and add anything new you find.”如何操作：“在 TASKS.md 中维护一份检查清单。每完成一项就勾选，并添加你发现的任何新任务。”

3. CHECKING THE RESULT3. 检查结果

Read what it needs from you first先阅读它需要你处理的内容

What to do. When a long run ends, look first for anything Claude is waiting on you for, like a decision it left open or a change it wants you to approve. Then read the rest of Claude’s summary.怎么做：当长任务结束时，首先查看 Claude 是否在等待你的反馈，例如它留下的待定决策或希望你批准的更改。然后再阅读 Claude 的其余总结。

Why it matters on Opus 5.5. Opus 5.5 reports on its work more clearly than Opus 5. Its updates and its final summary say what it did, what it found, and what it needs from you, in plain language.为什么这对 Opus 5.5 很重要：Opus 5.5 对工作进展的汇报比 Opus 5 更清晰。它的更新和最终总结会用简洁的语言说明它做了什么、发现了什么以及需要你做什么。

How. To change the summary’s format, say so in CLAUDE.md, for example, “End every run with three headings: Blocked on me, Changed, Found.”如何操作：若要更改总结的格式，请在 CLAUDE.md 中说明，例如：“在每次运行结束时，使用三个标题进行总结：待我处理、已更改、已发现。”

Ask it to review the code要求它进行代码审查

What to do. Ask Opus 5.5 to review a diff or a pull request before a person does.怎么做：在人工介入之前，要求 Opus 5.5 对差异（diff）或拉取请求（pull request）进行审查。

Why it matters on Opus 5.5. One early tester said Opus 5.5 at its lowest effort caught more bugs than Opus 5 at high effort, with fewer false alarms. It also explains its changes in plain language, so its pull request descriptions are easier to review.为什么这对 Opus 5.5 很重要：一位早期测试人员表示，Opus 5.5 即便在最低努力模式下，发现的漏洞也比 Opus 5 在高努力模式下更多，且误报更少。它还能用通俗易懂的语言解释其更改，因此其拉取请求描述更容易审查。

How. Feed this prompt to Claude:如何操作：向 Claude 输入此提示词：

Review the diff on this branch against main.
List only problems you'd block the merge for. For each one, give the file and line, why it's wrong, and how to show it fails.

Ask it to mark what it couldn’t confirm要求它标记无法确认的内容

What to do. For research and analysis, ask it to say what it couldn’t find or couldn’t check.怎么做：在研究和分析任务中，要求它列出无法找到或无法验证的内容。

Why it matters on Opus 5.5. “I couldn’t find this” is worth reading, and asking for it makes it easy to find.为什么这对 Opus 5.5 很重要：“我无法找到这个”是非常有价值的信息，要求它明确列出可以让你更容易发现问题。

How. Add “Mark anything you couldn’t confirm, and say where you looked” to the request. This works in a Claude research report and in Claude Code.如何操作：在请求中添加“标记出你无法确认的任何内容，并说明你的检索范围”。这在 Claude 研究报告和 Claude Code 中均有效。

4. IN CLAUDE APPS4. 在 CLAUDE 应用中

First, check that the model picker says Opus 5.5.首先，请确认模型选择器显示的是 Opus 5.5。

Share the chart or screenshot itself直接分享图表或截图

What to do. Attach the chart, diagram, screenshot, or slide. Don’t retype the numbers.怎么做：直接上传图表、示意图、截图或幻灯片。不要手动重新输入数据。

Why it matters on Opus 5.5. Opus 5.5 reads charts, diagrams, and screenshots more accurately than Opus 5, and it needs no extra steps to do it. It’s also better at meaning that depends on where things are in the image: which boxes an arrow connects, what changed between two versions of a diagram, or when a meeting starts and ends in a calendar screenshot.为什么这对 Opus 5.5 很重要：Opus 5.5 读取图表、示意图和截图的准确性高于 Opus 5，且无需额外步骤。它还更擅长理解依赖于图像布局的含义：比如箭头连接了哪些框，两版示意图之间有何变化，或者日历截图中会议的起止时间。

How. Attach the image and ask a specific question: “Which of these services call the billing API directly?”如何操作：上传图片并提出具体问题：“这些服务中，哪些直接调用了计费 API？”

Ask it to check a long document要求它检查长文档

What to do. Give it a long plan, report, or deck, and ask it to find mistakes.怎么做：提供一份长计划、报告或演示文稿，要求它找出其中的错误。

Why it matters on Opus 5.5. Opus 5.5 pays more attention to detail than prior Opus models. In our testing, it caught a date that fell on the wrong weekday in a long planning thread, and a chart that didn’t match the numbers in a deck.为什么这对 Opus 5.5 很重要：Opus 5.5 比以往的 Opus 模型更注重细节。在我们的测试中，它发现了一个长规划讨论串中日期对应星期错误的现象，以及演示文稿中图表与数据不符的情况。

How. Submit the prompt: “Check this deck for anything that contradicts itself: numbers, dates and names. Quote each problem and say where it is.”如何操作：提交提示词：“检查这份演示文稿中是否存在自相矛盾的内容：包括数字、日期和名称。引用每个问题并说明其位置。”

Ask for the finished file要求提供最终文件

What to do. When you want a spreadsheet or a document, ask for the file, not an outline.怎么做：当你需要电子表格或文档时，要求它生成文件，而不是大纲。

Why it matters on Opus 5.5. The spreadsheets and documents Opus 5.5 makes need less editing than Opus 5’s before you share them.为什么这对 Opus 5.5 很重要：Opus 5.5 生成的电子表格和文档在分享前所需的修改比 Opus 5 少得多。

How. “Make this a spreadsheet I can share: one row per vendor, with columns for cost, contract end date and owner.”如何操作：“将其制作成我可以分享的电子表格：每行对应一个供应商，包含成本、合同结束日期和所有者列。”

In a project, say when answers are settled在项目中，说明何时答案已确定

What to do. If follow-up questions in a long chat feel slow, add an instruction that earlier answers are settled.怎么做：如果长对话中的后续问题响应变慢，添加一条指令，说明之前的答案已经确定。

Why it matters on Opus 5.5. In a long chat, Opus 5.5 sometimes goes back over an earlier answer while it thinks about a short follow-up. That slows the reply.为什么这对 Opus 5.5 很重要：在长对话中，Opus 5.5 有时会在思考后续简短问题时回溯之前的答案。这会拖慢回复速度。

How. Add this to the project’s instructions:如何操作：将此内容添加到项目的指令中：

Once you have answered something, treat that answer as done. Focus on what I'm asking now, and don't go back over an earlier answer unless I ask about it or point out a problem with it.

Leave it out of projects for long analysis, where a later step can show a mistake in an earlier one.对于需要后续步骤可能推翻前期结论的长分析任务，请不要添加此项。

5. WHEN A MESSAGE IS FLAGGED5. 当消息被标记（Flagged）时

Opus 5.5 is the first Opus model to launch with Fable-level bio and cyber safeguards. In Claude apps and Claude Code, most flagged messages move to an older model, and your work goes on there. Finding security vulnerabilities in source code is allowed, and everyday health and educational questions should still work. These safeguards can sometimes flag legitimate work, and we’re tuning them to cut down on incorrect flags. If you’re switched, here’s what you’ll see and what to do.Opus 5.5 是首个在发布时即具备 Fable 级别生物和网络安全防护的模型。在 Claude 应用和 Claude Code 中，大多数被标记的消息会自动切换至旧模型，你的工作会在该模型中继续。在源代码中查找安全漏洞是允许的，日常的健康和教育类问题也应能正常工作。这些防护措施有时可能会误标合法工作，我们正在对其进行调整以减少误报。如果你被切换了模型，以下是你会看到的情况及应对方法。

In Claude apps在 Claude 应用中

What you see. A notice that starts with “Switched to” and the name of an older model. Claude answers on that model, and the chat stays on it.你会看到：一条以“已切换至”开头的通知，后跟旧模型的名称。Claude 会使用该模型进行回答，且后续对话将保持在该模型中。

What to do.怎么做：

• To go back to Opus 5.5, choose it in the model picker. If the earlier message is still in the chat, it may be flagged again. Starting a new chat avoids that.要切回 Opus 5.5，请在模型选择器中手动选择。如果之前的消息仍留在对话中，它可能会再次被标记。开启新对话可以避免这种情况。
• To be asked first, go to Settings, then Capabilities, and turn off “Switch models when a message is flagged.” You’ll see a “paused” card with your options.若要先询问再切换，请进入“设置” -> “功能”，关闭“当消息被标记时切换模型”。届时你会看到一个带有选项的“已暂停”卡片。

The check covers everything in the conversation, including files and search results. So a flag can come from earlier content, not only your last message.检查范围涵盖对话中的所有内容，包括文件和搜索结果。因此，标记可能源于之前的内容，而不仅仅是你的最后一条消息。

In Claude Code在 Claude Code 中

What you see. A notice that names the older model. The session continues on that model.你会看到：一条标明旧模型名称的通知。会话将继续在该模型上进行。

What to do.

• Run /model to switch back.运行 /model 以切回。
• Press Esc twice to edit your last message and try again.按两次 Esc 键以编辑最后一条消息并重试。
• To be asked first, run /config and change “Switch models when a message is flagged.”若要先询问再切换，请运行 /config 并更改“当消息被标记时切换模型”的设置。
• Run /feedback if the flag was wrong.如果标记有误，请运行 /feedback。

Don’t ask it to show its reasoning in the reply不要要求它在回复中展示其推理过程

What to do. Remove requests to reproduce its internal reasoning in the reply from your prompts and instructions.怎么做：从你的提示词和指令中移除要求其在回复中重现内部推理过程的请求。

Why it matters on Opus 5.5. A request to reproduce its internal reasoning in the reply can be declined. It’s one of the flag categories.为什么这对 Opus 5.5 很重要：要求重现内部推理过程的请求可能会被拒绝，这是标记类别之一。

How. Ask Claude for what you need instead, for example, “Explain why you chose this approach in three sentences.”如何操作：改为向 Claude 索取你需要的结果，例如：“用三句话解释你为什么选择这种方法。”

6. SPEED6. 速度

Turn on fast mode when you’re waiting on each reply当你需要等待每个回复时，开启快速模式（fast mode）

What to do. In Claude Code, use fast mode for back-and-forth work, where you read each reply before you send the next message.怎么做：在 Claude Code 中，对于需要反复交互的任务（即你在发送下一条消息前会先阅读上一个回复），使用快速模式。

Why it matters on Opus 5.5. Fast mode is available for Opus 5.5 at launch as a research preview. You get the same model, and the text arrives sooner. It needs extra usage turned on, and it costs more per token than standard mode.为什么这对 Opus 5.5 很重要：快速模式作为研究预览版随 Opus 5.5 同步推出。你使用的是同一个模型，但文本生成速度更快。它需要额外开启使用权限，且每个 Token 的成本高于标准模式。

How. Type /fast into Claude.如何操作：在 Claude 中输入 /fast。

YOUR OPUS 5.5 CHECKLIST你的 OPUS 5.5 检查清单

Run through this before your next long task.在进行下一个长任务前，请过一遍此清单。

Asking提问时

• The task says what “done” looks like任务已明确“完成”的标准
• No “think hard” lines in prompts or saved instructions提示词或保存的指令中没有“深度思考”之类的语句
• Design requests list the styles to leave out设计请求中列出了需要排除的样式
• Charts and screenshots are attached, not retyped直接上传图表和截图，而非重新输入

Long runs in Claude Code在 Claude Code 中执行长任务

• CLAUDE.md says when to stop and when to keep going, and to stop before anything destructiveCLAUDE.md 中明确了何时停止、何时继续，以及在执行破坏性操作前停止
• Permission prompts are still on for destructive commands对于破坏性命令，权限确认提示仍处于开启状态
• Large audits and migrations are split across subagents大型审计和迁移任务已拆分给子智能体
• The task list is kept in a file任务列表保存在文件中

Checking检查结果时

• The “needs from you” part of the report is read first已优先阅读报告中“需要你处理”的部分
• A review pass runs before a person reviews在人工审查前已运行审查流程
• Research answers mark what couldn’t be confirmed研究回答中已标记无法确认的内容

Flags关于标记（Flags）

• You know how to switch back: the model picker, or /model知道如何切回模型：通过模型选择器或 /model 命令
• “Switch models when a message is flagged” is set the way you want已按照你的偏好设置了“当消息被标记时切换模型”选项

Start building with Opus 5.5!开始使用 Opus 5.5 进行构建吧！

With thanks to Molly Vorwerck for reviewing.感谢 Molly Vorwerck 的审阅。