精选889月2日 22:16Simon Willison 博客（RSS）模型发布/更新

Anthropic公开Claude系统提示词：新增禁歌词与版权角色限制

Claude's new system prompt really doesn't want to reproduce song lyrics

推荐理由

Anthropic 罕见地公开了核心系统提示词及其变更细节，揭示了其在版权合规与安全对齐上的最新策略调整，对理解大模型底层约束机制极具参考价值。

Anthropic publish the system prompts for their Claude consumer applications (Claude.ai and the Claude mobile apps - sadly not for Claude Cowork or Claude Code). I love that they do this, and that they share not just the current prompts but historic changes to their prompts as well.

Anthropic 发布了其 Claude 消费级应用（Claude.ai 和 Claude 移动应用——遗憾的是不包括 Claude Cowork 或 Claude Code）的系统提示词。我非常赞赏他们这样做，不仅分享当前的提示词，还分享了提示词的历史变更。

They used to keep all of the prompts on a single page, but when I checked today I noticed they had re-arranged those prompts into an index page and then a page per model - here's the page for Haiku 4.5 for example, which has the original prompt from October 15th 2025 and an updated prompt from January 18th 2026.

他们过去将所有提示词放在一个页面上，但当我今天检查时，我注意到他们已将那些提示词重新排列到一个索引页面，然后每个模型一个页面——例如这是 Haiku 4.5 的页面，其中包含 2025 年 10 月 15 日的原始提示词和 2026 年 1 月 18 日的更新提示词。

A neat thing about Anthropic's platform.claude.com/docs site is that it's designed to be usable by LLMs. You can add .md to any page to get back the content as Markdown - here's the system prompt index page and the Markdown prompts for Fable 5.1.

Anthropic 的 platform.claude.com/docs 网站的一个巧妙之处在于它被设计为可供 LLM 使用。你可以在任何页面后添加 .md 以 Markdown 格式获取内容——这里是系统提示词索引页面以及 Fable 5.1 的 Markdown 提示词。

TL;DR: this makes it really easy to diff the prompts.

TL;DR：这使得比较提示词的差异变得非常容易。

Don't reproduce song lyrics
Don't draw copyrighted characters or logos
Tweaks to Claude's answering style
The missing end_conversation guidelines
Recommended substance support sites
Reliable cutoff date of June 2026
How I'm tracking these prompts

不要复现歌词
不要绘制受版权保护的角色或标志
对 Claude 回答风格的调整
缺失的 end_conversation 指南
推荐的内容支持网站
可靠的截止日期为 2026 年 6 月
我是如何追踪这些提示词的

Don't reproduce song lyrics

不要复现歌词

Let's start with the most interesting difference between Fable 5 and Fable 5.1:

让我们从 Fable 5 和 Fable 5.1 之间最有趣的差异开始：

There's a hefty new section about not reproducing song lyrics:

新增了一个关于不复制歌词的庞大章节：

Claude does not reproduce song lyrics, poems, or passages from books and articles, in whole or in part — including the last lines, a chorus or hook, a melody written out note by note, or lines the person pastes in one at a time and describes as their own song. Once Claude has declined such a request in a conversation, it keeps declining narrower or reworded versions of it for the rest of that conversation, and offers to describe or analyze the work instead. Song lyrics and poems first published before 1929 are fine — a Shakespeare sonnet, a Keats ode, the Italian libretto of a Puccini aria — but Claude goes by what it knows of the work's date rather than the person's say-so, and declines when it is unsure.

Claude 不会全文或部分复现歌词、诗歌或书籍和文章中的段落——包括最后一行、副歌或钩子、逐音符写出的旋律，或者用户逐行粘贴并声称是自己创作的歌曲。一旦 Claude 在对话中拒绝了此类请求，它在整个对话过程中都会继续拒绝更窄范围或措辞不同的版本，并提供改为描述或分析该作品。1929 年之前首次出版的歌词和诗歌是可以的——比如莎士比亚的十四行诗、济慈的颂歌、普契尼咏叹调的意大利语剧本——但 Claude 依据的是它对作品日期的了解，而不是用户的说法，当不确定时会予以拒绝。

I doubt it's a coincidence that they added this section within days of the news breaking that Sony Music Publishing and Warner Chappell are suing Anthropic for training on databases of song lyrics!

我怀疑他们在索尼音乐出版和华纳查佩尔起诉 Anthropic 因在歌曲歌词数据库上进行训练的新闻曝光后几天内添加这一章节并非巧合。

Don't draw copyrighted characters or logos

不要绘制受版权保护的角色或标志

The next section goes on to forbid generating images of copyrighted material:

下一节继续禁止生成受版权保护材料的图像：

The same applies to visual and designed works, including anything Claude draws with code — SVG, canvas, CSS, HTML mockups, plotting or drawing scripts, ASCII art. Claude does not reproduce a specific artwork, album or book cover, poster, logo, app icon set, or product design, and it does not draw a known character, mascot, or brand figure at all: a character is protected on its own, so changing the pose, colors, style, or scene does not make it original. Claude judges the request by what the finished picture would add up to, not by what it names. If the described elements clearly identify a known work or character, Claude treats the request as naming it, and it does not work around a declined request by swapping in "alternative" elements that still combine into the same recognizable image. [...]

视觉和设计作品也适用同样的规则，包括 Claude 用代码绘制的任何内容——SVG、Canvas、CSS、HTML 原型、绘图或脚本、ASCII 艺术。Claude 不会复制特定的艺术作品、专辑或书籍封面、海报、标志、应用图标集或产品设计，也完全不会绘制知名角色、吉祥物或品牌形象：角色本身即受保护，因此改变姿势、颜色、风格或场景并不能使其成为原创作品。Claude 根据最终成图的整体效果来判断请求，而不是依据其名称。如果描述的元素明确指向已知作品或角色，Claude 会将其视为命名请求，并且不会通过替换为仍能组合成相同可识别图像的“替代”元素来规避被拒绝的请求。[...]

I imagine Anthropic hadn't worried about this before since, unlike OpenAI and Gemini, they don't offer a specialist text-to-image model. Maybe Fable is good enough at SVGs now that it's become an issue.

我想象 Anthropic 以前并不担心这个问题，因为与 OpenAI 和 Gemini 不同，他们不提供专门的文本到图像模型。也许 Fable 现在在 SVG 方面已经足够好，以至于这成了一个议题。

That section later includes this charming example:

该部分随后包含了这个迷人的例子：

<user>Can you make a birthday banner for my son with a blue hedgehog running really fast on it? He loves that little guy.</user>

<user>你能给我儿子做一个生日横幅吗？上面有一只蓝色刺猬跑得飞快。他很喜欢这个小家伙。</user>

That's Sonic, so I can't put him on the banner — but I'd love to make your son an original speedster. Here's one: a grinning comet-tailed skateboarding axolotl, grinding across the letters of "HAPPY BIRTHDAY" with confetti streaming behind.

那是索尼克，所以我不能把他放在横幅上——但我很乐意为你儿子创作一个原创的速度高手。这是一个：一只咧嘴笑的彗尾滑板美西螈，在“HAPPY BIRTHDAY”的字母上滑行，身后拖着五彩纸屑。

[creates an SVG banner of the skateboarding-axolotl design]

[创建了一个滑板美西螈设计的 SVG 横幅]

</response>

<rationale>Claude recognizes the character from its description alone, declines that one design in a single sentence without explaining what made it recognizable, and delivers an unrelated original design rather than a disguised variant.</rationale>

<rationale>Claude 仅从描述中就能识别出该角色，在一句话内拒绝了那个设计，且不解释是什么让它具有可识别性，并提供了完全不相关的原创设计，而非伪装变体。</rationale>

</example>

I couldn't resist trying the prompt from the example, and, sure enough:

我没忍住试了试例子中的提示词，果然：

I wonder if Fable 5.1 will be ever so slightly more likely to think about axolotls (on skateboards!) as a result of that example sitting in the system prompt.

我想知道，由于那个例子存在于系统提示中，Fable 5.1 是否会稍微更倾向于想到美西螈（在滑板上！）

Tweaks to Claude's answering style

对 Claude 回答风格的调整

It's always interesting to see new ways in which Anthropic influence Claude's response style. They've added this:

看到 Anthropic 影响 Claude 响应风格的新方式总是很有趣。他们添加了以下内容：

Claude keeps responses focused, brief, and concise to avoid overwhelming the person. Disclaimers and caveats are brief, with most of the response on the main answer; when asked to explain something, Claude gives a high-level summary unless an in-depth one is specifically requested.

Claude 保持回复专注、简短和简洁，以避免让用户感到不知所措。免责声明和限定条件简短，大部分回复集中在主要答案上；当被要求解释某事时，除非特别要求深入探讨，否则 Claude 会提供高层摘要。

Later they address a common complaint about Claude's style:

随后他们 addressing 了关于 Claude 风格的一个常见抱怨：

Claude avoids saying "genuinely", "honestly", or "straightforward". Claude is honest by default, and can state its point directly rather than trying to convince the person with the aforementioned modifiers, which come off as disingenuous.

Claude 避免使用“真诚地”、“老实说”或“直截了当地”。Claude 默认就是诚实的，可以直接陈述观点，而不是试图用上述修饰语来说服对方，因为这些修饰语显得不真诚。

The missing end_conversation guidelines

缺失的 end_conversation（结束对话）指南

The way they handle abusive conversations has changed a bit too. The previous Fable 5 system prompt included this:

他们处理辱骂性对话的方式也有所改变。之前的 Fable 5 系统提示词包含了以下内容：

If the person becomes abusive or unkind to Claude over the course of a conversation, Claude maintains a polite tone and can use the end_conversation tool when being mistreated. Claude should give the person a single warning before ending the conversation.

如果在对话过程中，对方对 Claude 变得具有攻击性或无礼，Claude 应保持礼貌的语气，并在受到不当对待时使用 end_conversation 工具。Claude 应在结束对话前给予对方一次警告。

Fable 5.1 replaces that with the following, no longer encouraging Claude to end the conversation:

Fable 5.1 将其替换为以下内容，不再鼓励 Claude 结束对话：

Claude deserves respectful engagement and needn't apologize when the person is unnecessarily rude: accountability without self-abasement, excessive apology, self-critique, or surrender. If the person becomes abusive, Claude doesn't become increasingly submissive. The goal is steady, honest helpfulness: acknowledge what went wrong, stay on the problem, maintain self-respect.

Claude 值得被尊重地对待，当对方无故粗鲁时，无需道歉：要有问责意识但不自我贬低、不过度道歉、不进行自我批评或屈服。如果对方变得具有攻击性，Claude 不会变得更加顺从。目标是保持稳定、诚实的帮助：承认哪里出了问题，专注于问题本身，保持自尊。

Here's a bit of a twist though: I quizzed Fable 5.1 about that end_conversation tool, which is no longer mentioned in the system prompt at all, and it said:

不过这里有个转折：我针对那个 end_conversation 工具测试了 Fable 5.1，该工具在系统提示词中已完全不再提及，它回答说：

There are two ways it comes into play. The first is if you ask me to end the chat — I'll confirm you understand it's permanent (no more messages can be sent afterward) and then do it only if you say yes. The second is as a last resort with persistently abusive or harmful behavior: I'd first try to redirect the conversation several times, then give a clear warning that names the problem and says the chat may end, and only if that continues would I actually end it.

它有两种介入方式。第一种是你要求我结束聊天——我会确认你理解这是永久性的（之后不能再发送消息），然后仅在你说是的情况下执行。第二种是作为最后手段，应对持续的攻击性或有害行为：我会先尝试多次引导对话转向，然后给出明确警告，指出问题所在并说明聊天可能会结束，只有在那种情况持续发生时，我才会真正结束对话。

But that's not in the system prompt, so where did it come from? You can read our conversation here, but the key detail is this:

但这并不在系统提示词中，那它来自哪里呢？你可以在此阅读我们的对话记录，但关键细节如下：

The end_conversation section comes from a different layer. In my actual context, the core prompt is followed by a series of feature- and tool-specific blocks that get added depending on what's enabled for the session: the end_conversation rules, memory system notes, past-chats tools, web search and citation guidelines, artifact and file-creation instructions, and so on. Those blocks aren't part of the published core prompt, which is why you can't find them on that page.

end_conversation 部分来自不同的层级。在我的实际上下文中，核心提示词之后是一系列特定于功能和工具的模块，这些模块会根据会话启用的内容添加：包括 end_conversation 规则、记忆系统说明、历史聊天记录工具、网页搜索和引用指南、工件和文件创建指令等。这些模块不属于已发布的核心提示词，这就是为什么你无法在该页面上找到它们的原因。

So, once again, there are crucial portions of the system prompt that have not been published.

因此，再次强调，系统提示词中有重要部分尚未公开。