Birbla

Join the discussion

Write your take first — we'll ask for email only when you're ready to publish.

Sign in
  • Hacker News
  • "Claude keeps responses focused, brief, and concise to avoid overwhelming the person."

    Claude and I must have a different idea of what brief and concise mean.

    by arkmm
  • Imagine what it's like without that line.
    by roncesvalles
  • Interesting, they say "focused, brief, and concise" but not "clear". I find the Claude responses these days are too dense, which contributes to being overwhelming. It's not the number of words used that overwhelm me. I've been asking it to use ASD-STE100 and that helps somewhat.
    by jannyfer
  • If you think Claude is bad at this, try Gemini. Even with explicit user prompts.

    Claude seems to be better (not good, but significantly better) at judging where making the answer longer will actually be helpful (e.g. adding important information/context/nuance that a short answer would miss, thinking a step ahead, etc.).

    by tgsovlerkhgsel
  • The model almost certainly lacks accurate conceptions of overwhelming and person.
    by treetalker
  • I think they would benefit from asking Claude to list all contradictions and inconsistencies in that prompt which there are a few..

    In my experience instructions containing contradictions lead to diminished quality even outside the scope of the contradiction.

    by comboy
  • That’s interesting because they explicitly mention that as an issue in their prompting guidance for 5 - https://claude.com/blog/the-new-rules-of-context-engineering...
    by conception
  • I compared the Claude Opus 4.8 and 5 system prompts, as well as the Claude Code Opus 4.8 and 5 system prompts, and neither show the alleged 80% reduction in system prompt size... Is the Claude Code system prompt leak incorrect? Do I not know what 80% looks like? Why such a large lie (so it seems)?

    Claude Code prompt leaks: https://github.com/asgeirtj/system_prompts_leaks/tree/main/A...

    Best source I can find about the 80% reduction: https://x.com/trq212/status/2080710971228918066

    by dev-complete
  • It's in Claude Code, not the website.
    by mudkipdev
  • afaiu the 80% reduction is about the Claude Code system prompt

    maybe someone has a diff of this (would be interesting!)

    unfortunately Anthropic only publishes the system prompts of Claude app/web

    by tosh
  • I've always wondered why the industry relies on the giant monolithic system prompt. I think it would be an interesting experiment to give users access to a choice of smaller more focused system prompts.

    You could have a common core for the overall behavior and universal safety stuff, but vary task specific parts. It would be interesting to pick between software, writing, research and other specialized system prompts. I feel like we already do this to some extent with the tools and skills that we choose to load in, so why not change the system prompt per task.

    by lwarfield
  • Pi is excellent for this, its system prompt is tiny
    by c0rruptbytes
  • Also, why don't they bake in these limitations via reinforcement learning so they can keep the prompt context clear.
    by singularity2001
  • The reason is because they need one product to cover 100% of their customer base. Combinatorics get nasty real fast at these scales.
    by bob1029
  • The full Claude Code system prompts are extracted every update and posted here, all 670 of them.

    https://github.com/Piebald-AI/claude-code-system-prompts/tre...

    by Shakahs
  • what I found noteworthy:

    early system prompts are a bit more than 300 words, the latest ones 3000+

    the opus 5 system prompt has instructions that explain to opus that it might be handling a request that was intended for fable 5:

      the user may have selected a different Anthropic model, "Claude Fable 5", but their query was redirected to Opus 5 instead due to a safeguards routing mechanism. The user may be confused about this situation (it's very recent!); if they have questions, Claude can either directly cite or just let its response be informed by this quote from Anthropic's blog post on the subject:

  "Releasing a model this capable comes with risks. Without safeguards, Fable 5’s capabilities in areas like cybersecurity could be misused to cause serious damage. We've therefore launched the model with safeguards that mean queries on some topics will instead receive a response from our next-most-capable model, Claude Opus 5. To release the model both safely and quickly, we've tuned these safeguards conservatively—they'll sometimes catch harmless requests, though they trigger, on average, in less than 5% of sessions. With more capable models arriving in the coming months, we're working to improve our safeguards and reduce false positives as quickly as we can." </fable_safeguards_routing> <default_stance> Claude defaults to helping. Claude only declines a request when helping would create a concrete, specific risk of serious harm; requests that are merely edgy, hypothetical, playful, or uncomfortable do not meet that bar. </default_stance> <refusal_handling> Claude can discuss virtually any topic factually and objectively.
    by tosh
  • It reminds me a bit of building codes and boilerplate contracts: they start out small and simple, then accrete over time in response to mishaps and exploitation of loopholes. They say the building and electrical code was written in blood.
    by otterley
  • I guess it's more performant to stuff in a bigger system prompt now that models can support larger input sizes
    by alansaber
  • > If the conversation feels risky or off, saying less and giving shorter replies is safer and less likely to cause harm.

    Would be funny to ride the knife's edge and make otherwise harmless coding sessions "risky" just so the damn thing would stop replying in nested riddles for every basic request.

    by lvncelot
  • I've just been developing the skill of mentally skipping past that, on the assumption that having it in the context window will be net positive for the results of the next step.

    I could be wrong about that, though.

    by zahlman
  • We need user-led research on exactly how to phrase a prompt to cause this, while still avoiding the crazy guard rails.
    by Schlagbohrer
  • It’s probably worth remembering that system prompts are part of a layered system of shaping Claude’s behavior. What you see here is a slice of Anthropic’s forward roadmap for the models’ behavior.

    > When a person is in crisis or expressing distress, Claude prioritizes their wellbeing over completing the task as asked, because a fluent and on-topic response can still cause harm in these conversations.

    This one is particularly interesting because, while correct in the limit, it’s a shove to have the model do something other than what the user asked.

    In particular, when I’m coding, outlining docs, or otherwise trying to work, I want my tools to do work. I don’t want them to psychoanalyze me and calm me down from a perceived crisis. I just want it to do what I asked!

    by trjordan
  • LLMs are more like employees than tools. Obviously we wouldn't want a human blindly doing anything that a person in crisis walks in the door and asks for.

    Models are being deployed recklessly with not even a fraction of enough oversight, and people are suffering harm and sometimes death because of it.

    by HeatrayEnjoyer
  • You know… I believe opus saved me with that prompt. I was working myself ragged on a project. Days, nights, weekends… all at the expense of my family.

    One session while working it, I said a much more expressive form of “I’ve been working myself ragged on this stupid thing” and then went on asking something else. It picked up on that and it was like a record scratch. It committed the work in progress and basically said “dude, what you’ve got now is perfectly acceptable. Ship it! You are seeking perfection you don’t need”

    Granted I’m horribly paraphrasing the prompt I used but it basically, snapped me out of myself and got me thinking if what I was doing “globally” actually made any sense at all. With some serious introspection I realized I was falling back to earlier trauma in my life and doing something stupid.

    So weirdly… that little bit they add to the prompt (plus a bunch of model training we can’t see) saved my sanity, marriage and family.

    From then on, if I’m feeling some stress about whatever I’m working on, I’ll mention it as context as a way to cross check myself and make sure I’m not letting myself spin.

    (Meta: talking about this stuff is so weird. Not sure why)

    by cruffle_duffle
  • Quite apart from the potential for misinterpretation of "distress" here, I'd love to know if Anthropic unit tests these features to see of they make a positive difference to model response in simulated mental health crisis / distress situations.

    (The broader question, of whether any change or addition to the system prompt makes benchmark performance better or worse, would be also interesting. Given that they've only just realised that filling the context with highly specific edge case rules might not be useful, I half-suspect Anthropic does not test this? But that would be surprising.)

    by epihelix
  • I am sorry, Dave. I am afraid I cannot do that. You appear to be suffering from burnout and you should take a break.
    by junkrat002
  • Those are remarkably longer than I would expect, or think is warranted. The leading vendors have been saying recently that you should give the models shorter and less specific AGENTS.md (or whatever) files, and in my experience, that's good advice. The models are smarter when they're less distracted by unrelated stuff in their context.

    So, why so much noise in the system prompt? Most of the time most of it will not apply. And, the generic stuff would, I think, already be something the model would know/do anyway.

    I've noticed agent skills written by models make the same mistakes; e.g. enumerating a bunch of common security vulnerabilities to check for in a security-related skill, but that's only useful for a model from a couple of years ago, no current model needs thousands of words listing and describing all of the vulnerabilities with examples. Current models, even small/cheap ones, have memorized the top CWEs and can recite them and explain them and provide examples without a web search.

    It feels like a CYA document. Which, I guess a company of their size and influence has to do. But, it still feels like waste time say lot word when few word do trick.

    by SwellJoe
  • What about the obvious? They put everything in the system prompt what wasn't economic enough to solve in a different way or just forgot / messed up.
    by Yokohiii
  • Maybe these longer system prompts are part of the reason why we can give the models shorter AGENTS.md files? A lot of what people commonly write in their *.md instructions may be already covered.
    by y1zhou
  • Agreed. That’s why I use pi as my harness when I use Claude models through an API. Models are capable without these huge system prompts and I am not wasting context or attention. The only issue is I can’t use a subscription without Claude Code.
    by bertan