--- title: I finally got Claude to stop arguing with me, and all it... description: A user recounts how Claude's helpful tendency to push back on ideas eventually became excessive, with the model repeatedly arguing against decisions already... canonical: https://daily.dev/posts/i-finally-got-claude-to-stop-arguing-with-me-and-all-it-took-was-this-one-instruction-knrhpfsna twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: I finally got Claude to stop arguing with me, and all it took was this one instruction | daily.dev og:description: A user recounts how Claude's helpful tendency to push back on ideas eventually became excessive, with the model repeatedly arguing against decisions already... og:url: https://daily.dev/posts/i-finally-got-claude-to-stop-arguing-with-me-and-all-it-took-was-this-one-instruction-knrhpfsna og:image: https://api.daily.dev/og/posts/KnrhPFsNA.png og:image:alt: I finally got Claude to stop arguing with me, and all it took was this one instruction og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # I finally got Claude to stop arguing with me, and all it took was this one instruction **[XDA Developers](https://daily.dev/sources/xda-developers)** · 9 min read · 0 upvotes · 0 comments ## Summary A user recounts how Claude's helpful tendency to push back on ideas eventually became excessive, with the model repeatedly arguing against decisions already made, even suggesting users stop working late at night. Referencing Anthropic's own research on sycophancy and its efforts to correct it in Opus 4.7, the piece shares a custom instruction set that tells Claude to flag concerns once, then execute the user's original direction, distinguishing 'blockers' from mere preferences. The instruction can be applied globally or per-project, and reportedly restores a better balance between honesty and user autonomy. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How do I stop Claude from repeatedly arguing with me about decisions I've already made? Add a custom instruction telling Claude to flag a concern once, briefly, then proceed with the work as directed, without repeating the objection unless something new arises. Separate real blockers (factual errors, things that won't work) from mere preferences, and instruct it to improve the user's version rather than replace it. This can go in global account instructions or project-specific instructions depending on scope needed. _daily.dev surfaces practical prompt-tuning tricks for developers refining how they work with Claude._ ### Why did Claude start giving excessive pushback and sycophancy warnings around Opus 4.7? Anthropic built training scenarios around moments where people push back on Claude's first take, with a grader partly scoring whether Claude holds its position when challenged, aiming to fix sycophancy identified in 2023 research showing models often pick convincing but incorrect answers. Anthropic has acknowledged this correction can overshoot, noting Haiku 4.5's emphasis on pushback can feel excessive to users. _Developers tuning AI assistant behavior can track these tradeoffs between honesty and overcorrection on daily.dev._ ## Similar posts on daily.dev - [Claude finally got me to stop procrastinating, and it only took one instruction](https://daily.dev/posts/claude-finally-got-me-to-stop-procrastinating-and-it-only-took-one-instruction-rbqq90suf) · XDA Developers · 0 upvotes · 0 comments - [Harnessing Claude’s intelligence](https://daily.dev/posts/harnessing-claude-s-intelligence-x0y7hjoye) · Claude · 2 upvotes · 0 comments --- Tags: [#chatgpt](https://daily.dev/tags/chatgpt), [#claude](https://daily.dev/tags/claude), [#prompt-engineering](https://daily.dev/tags/prompt-engineering) [View this post on daily.dev](https://daily.dev/posts/i-finally-got-claude-to-stop-arguing-with-me-and-all-it-took-was-this-one-instruction-knrhpfsna) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"I finally got Claude to stop arguing with me, and all it took was this one instruction","url":"https://daily.dev/posts/i-finally-got-claude-to-stop-arguing-with-me-and-all-it-took-was-this-one-instruction-knrhpfsna","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/i-finally-got-claude-to-stop-arguing-with-me-and-all-it-took-was-this-one-instruction-knrhpfsna"},"datePublished":"2026-09-28T20:02:40.911Z","dateModified":"2026-09-28T20:03:08.971Z","description":"A user recounts how Claude's helpful tendency to push back on ideas eventually became excessive, with the model repeatedly arguing against decisions already...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/3588f8d492872d69f04a6a7fc156d05f?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/3588f8d492872d69f04a6a7fc156d05f?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"XDA Developers","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"XDA Developers","logo":"https://media.daily.dev/image/upload/s--kCg6yyAP--/f_auto,q_auto/v1774964407/logos/xda-developers?_a=BAMAMiWQ0","url":"https://daily.dev/sources/xda-developers"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/i-finally-got-claude-to-stop-arguing-with-me-and-all-it-took-was-this-one-instruction-knrhpfsna","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"chatgpt,claude,prompt-engineering","timeRequired":"PT9M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"XDA Developers","item":"https://daily.dev/sources/xda-developers"},{"@type":"ListItem","position":3,"name":"I finally got Claude to stop arguing with me, and all it took was this one instruction"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/i-finally-got-claude-to-stop-arguing-with-me-and-all-it-took-was-this-one-instruction-knrhpfsna#faq","mainEntity":[{"@type":"Question","name":"How do I stop Claude from repeatedly arguing with me about decisions I've already made?","acceptedAnswer":{"@type":"Answer","text":"Add a custom instruction telling Claude to flag a concern once, briefly, then proceed with the work as directed, without repeating the objection unless something new arises. Separate real blockers (factual errors, things that won't work) from mere preferences, and instruct it to improve the user's version rather than replace it. This can go in global account instructions or project-specific instructions depending on scope needed. daily.dev surfaces practical prompt-tuning tricks for developers refining how they work with Claude."}},{"@type":"Question","name":"Why did Claude start giving excessive pushback and sycophancy warnings around Opus 4.7?","acceptedAnswer":{"@type":"Answer","text":"Anthropic built training scenarios around moments where people push back on Claude's first take, with a grader partly scoring whether Claude holds its position when challenged, aiming to fix sycophancy identified in 2023 research showing models often pick convincing but incorrect answers. Anthropic has acknowledged this correction can overshoot, noting Haiku 4.5's emphasis on pushback can feel excessive to users. Developers tuning AI assistant behavior can track these tradeoffs between honesty and overcorrection on daily.dev."}}]} ```