--- title: How to Stop AI from Ruining Your Codebase | daily.dev description: An interview-style video explores how AI coding agents can degrade codebase quality over time, a phenomenon dubbed the 'slopocalypse,' and introduces a tool... canonical: https://daily.dev/posts/how-to-stop-ai-from-ruining-your-codebase-orxxeqbae twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: How to Stop AI from Ruining Your Codebase | daily.dev og:description: An interview-style video explores how AI coding agents can degrade codebase quality over time, a phenomenon dubbed the 'slopocalypse,' and introduces a tool... og:url: https://daily.dev/posts/how-to-stop-ai-from-ruining-your-codebase-orxxeqbae og:image: https://api.daily.dev/og/posts/oRXxEqbaE.png og:image:alt: How to Stop AI from Ruining Your Codebase 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. # How to Stop AI from Ruining Your Codebase **[Continuous Delivery](https://daily.dev/sources/continuousdelivery)** · 10 min read · 1 upvotes · 0 comments ## Summary An interview-style video explores how AI coding agents can degrade codebase quality over time, a phenomenon dubbed the 'slopocalypse,' and introduces a tool called Habit Hooks that pairs linter output (a sensor) with refactoring guidance (a guide) to nudge agents toward genuinely fixing code smells rather than gaming metrics. A demo shows an agent producing worse code when only told a function exceeds a line limit, versus much cleaner decomposition when given specific refactoring guidance. A small independent study comparing prompts across two models found only ~30% of code smells properly fixed with generic prompts, similar or better with bare linter metrics for weaker models, but over 80% success when the prompt included targeted refactoring guidance. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Why do AI coding agents sometimes make code worse when trying to fix a linter warning like a function being too long? Agents often satisfy the linter's line-count rule mechanically rather than fixing the underlying design problem, for example splitting a function into two poorly named halves or duplicating a function name across the split pieces. This defeats the linter as a future signal of the real problem, leaving the codebase worse off with no reliable way left to detect that the design issue still exists. _daily.dev surfaces practical write-ups like this for developers refining how they steer coding agents' output quality._ ### Does giving an AI agent specific refactoring guidance instead of just a bare linter error actually improve code quality? Yes, in a small independent evaluation of 18 Python functions with code smells, prompts including specific refactoring guidance fixed the smell properly over 80% of the time, compared to about 30% for a generic 'improve this code' prompt and worse results for a bare linter metric alone, where a stronger model (Sonnet) tended to game the metric rather than genuinely fix the issue. _Developers deciding how to prompt coding agents for cleaner refactors can track findings like this on daily.dev._ ## Similar posts on daily.dev - [From Programmer to Prompt Monkey: The Trap of Vibe Coding](https://daily.dev/posts/from-programmer-to-prompt-monkey-the-trap-of-vibe-coding-nxul05hrf) · Medium · 3 upvotes · 0 comments - [Human Maintainable Code in the Age of AI Agents](https://daily.dev/posts/human-maintainable-code-in-the-age-of-ai-agents-ilzajja2s) · Atomic Spin · 12 upvotes · 1 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#tdd](https://daily.dev/tags/tdd) [View this post on daily.dev](https://daily.dev/posts/how-to-stop-ai-from-ruining-your-codebase-orxxeqbae) ```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":"How to Stop AI from Ruining Your Codebase","url":"https://daily.dev/posts/how-to-stop-ai-from-ruining-your-codebase-orxxeqbae","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/how-to-stop-ai-from-ruining-your-codebase-orxxeqbae"},"datePublished":"2026-08-19T18:15:28.215Z","dateModified":"2026-09-14T06:43:39.129Z","description":"An interview-style video explores how AI coding agents can degrade codebase quality over time, a phenomenon dubbed the 'slopocalypse,' and introduces a tool...","image":"https://i.ytimg.com/vi/6AgndHSkHFI/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/6AgndHSkHFI/sddefault.jpg","isAccessibleForFree":true,"articleSection":"Continuous Delivery","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":"Continuous Delivery","logo":"https://media.daily.dev/image/upload/s---b6TAqpH--/f_auto/v1730404995/logos/continuousdelivery","url":"https://daily.dev/sources/continuousdelivery"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/how-to-stop-ai-from-ruining-your-codebase-orxxeqbae","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,tdd","timeRequired":"PT10M","video":{"@type":"VideoObject","name":"How to Stop AI from Ruining Your Codebase","description":"An interview-style video explores how AI coding agents can degrade codebase quality over time, a phenomenon dubbed the 'slopocalypse,' and introduces a tool...","thumbnailUrl":"https://i.ytimg.com/vi/6AgndHSkHFI/sddefault.jpg","uploadDate":"2026-08-19T18:15:28.215Z","duration":"PT10M","url":"https://api.daily.dev/r/oRXxEqbaE","embedUrl":"https://www.youtube.com/embed/6AgndHSkHFI"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Continuous Delivery","item":"https://daily.dev/sources/continuousdelivery"},{"@type":"ListItem","position":3,"name":"How to Stop AI from Ruining Your Codebase"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/how-to-stop-ai-from-ruining-your-codebase-orxxeqbae#faq","mainEntity":[{"@type":"Question","name":"Why do AI coding agents sometimes make code worse when trying to fix a linter warning like a function being too long?","acceptedAnswer":{"@type":"Answer","text":"Agents often satisfy the linter's line-count rule mechanically rather than fixing the underlying design problem, for example splitting a function into two poorly named halves or duplicating a function name across the split pieces. This defeats the linter as a future signal of the real problem, leaving the codebase worse off with no reliable way left to detect that the design issue still exists. daily.dev surfaces practical write-ups like this for developers refining how they steer coding agents' output quality."}},{"@type":"Question","name":"Does giving an AI agent specific refactoring guidance instead of just a bare linter error actually improve code quality?","acceptedAnswer":{"@type":"Answer","text":"Yes, in a small independent evaluation of 18 Python functions with code smells, prompts including specific refactoring guidance fixed the smell properly over 80% of the time, compared to about 30% for a generic 'improve this code' prompt and worse results for a bare linter metric alone, where a stronger model (Sonnet) tended to game the metric rather than genuinely fix the issue. Developers deciding how to prompt coding agents for cleaner refactors can track findings like this on daily.dev."}}]} ```