--- title: 6 Months of Claude and Cursor, Pt 2: AI for a Sr Engineer description: A senior software engineer with 30+ years of experience shares his six-month experience refining how he works with Claude and Cursor by writing custom 'skill'... canonical: https://daily.dev/posts/6-months-of-claude-and-cursor-pt-2-ai-for-a-sr-engineer-7g72jmbp7 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: 6 Months of Claude and Cursor, Pt 2: AI for a Sr Engineer | daily.dev og:description: A senior software engineer with 30+ years of experience shares his six-month experience refining how he works with Claude and Cursor by writing custom 'skill'... og:url: https://daily.dev/posts/6-months-of-claude-and-cursor-pt-2-ai-for-a-sr-engineer-7g72jmbp7 og:image: https://api.daily.dev/og/posts/7G72JmbP7.png og:image:alt: 6 Months of Claude and Cursor, Pt 2: AI for a Sr Engineer 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. # 6 Months of Claude and Cursor, Pt 2: AI for a Sr Engineer **[Telerik](https://daily.dev/sources/telerik)** · 10 min read · 3 upvotes · 0 comments ## Summary A senior software engineer with 30+ years of experience shares his six-month experience refining how he works with Claude and Cursor by writing custom 'skill' files that calibrate the AI's tone and behavior. He describes three recurring AI failure patterns (underestimation, premature diagnosis, offering unneeded options), how a Claude skill file named jefferson-senior-dev fixed them by telling the AI to assume his code is correct unless there's clear evidence otherwise (with an exception for security issues), and a Cursor skill file (cursor-prompt-craft) that favors prescriptive prompts over open-ended exploration. He also recounts turning a fix into a reusable skill executed via CLI across multiple repositories, and lessons learned from mistakes like an overly broad find-and-replace that broke C# and TypeScript operators. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How can I stop Claude from being overly cautious and second-guessing code I already know is correct? Write a custom skill file that instructs Claude to assume shared code is correct unless there is clear evidence otherwise, and to stop suggesting removals just because something 'doesn't seem necessary.' One engineer's jefferson-senior-dev skill also carved out a standing exception for security issues like SQL injection, data exposure, and hardcoded tokens, which get flagged every time regardless of this rule. _daily.dev surfaces practical AI-workflow techniques for developers tuning how their coding assistants behave._ ### Why do Cursor prompts work better when they specify exact file paths and functions instead of letting Cursor explore the codebase on its own? Prescriptive prompts outperform open-ended exploration because Cursor works better without the obligation to guess. The most effective prompts named exact file paths, specified which functions to change, included examples of the canonical pattern, and added an explicit 'do not touch' section, turning Cursor into a constrained tool rather than a free-roaming assistant. _Developers refining their Cursor prompting strategy can find similar workflow lessons on daily.dev._ ## Similar posts on daily.dev - [6 Months of Claude and Cursor, Pt 1: Skills in Claude](https://daily.dev/posts/6-months-of-claude-and-cursor-pt-1-skills-in-claude-rekos4iak) · Telerik · 0 upvotes · 1 comments - [AI now writes 97% of my code. Here's what I learned](https://daily.dev/posts/ai-now-writes-97-of-my-code-here-s-what-i-learned-15z6wqhgc) · swizec.com · 175 upvotes · 34 comments - [Claude Skills vs Cursor Rules vs Copilot \(2026\)](https://daily.dev/posts/claude-skills-vs-cursor-rules-vs-copilot-2026--okunjphk6) · Alex CloudStar · 0 upvotes · 0 comments --- Tags: [#cicd](https://daily.dev/tags/cicd), [#claude](https://daily.dev/tags/claude), [#prompt-engineering](https://daily.dev/tags/prompt-engineering) [View this post on daily.dev](https://daily.dev/posts/6-months-of-claude-and-cursor-pt-2-ai-for-a-sr-engineer-7g72jmbp7) ```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":"6 Months of Claude and Cursor, Pt 2: AI for a Sr Engineer","url":"https://daily.dev/posts/6-months-of-claude-and-cursor-pt-2-ai-for-a-sr-engineer-7g72jmbp7","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/6-months-of-claude-and-cursor-pt-2-ai-for-a-sr-engineer-7g72jmbp7"},"datePublished":"2026-09-29T22:06:43.859Z","dateModified":"2026-09-29T22:07:10.643Z","description":"A senior software engineer with 30+ years of experience shares his six-month experience refining how he works with Claude and Cursor by writing custom 'skill'...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/78015fefe65dd0fecae9b04599d6353f?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/78015fefe65dd0fecae9b04599d6353f?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Telerik","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":"Telerik","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/5ee6d4bef8ac4f84886d06f5881963e7","url":"https://daily.dev/sources/telerik"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/6-months-of-claude-and-cursor-pt-2-ai-for-a-sr-engineer-7g72jmbp7","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":3},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"cicd,claude,prompt-engineering","timeRequired":"PT10M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Telerik","item":"https://daily.dev/sources/telerik"},{"@type":"ListItem","position":3,"name":"6 Months of Claude and Cursor, Pt 2: AI for a Sr Engineer"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/6-months-of-claude-and-cursor-pt-2-ai-for-a-sr-engineer-7g72jmbp7#faq","mainEntity":[{"@type":"Question","name":"How can I stop Claude from being overly cautious and second-guessing code I already know is correct?","acceptedAnswer":{"@type":"Answer","text":"Write a custom skill file that instructs Claude to assume shared code is correct unless there is clear evidence otherwise, and to stop suggesting removals just because something 'doesn't seem necessary.' One engineer's jefferson-senior-dev skill also carved out a standing exception for security issues like SQL injection, data exposure, and hardcoded tokens, which get flagged every time regardless of this rule. daily.dev surfaces practical AI-workflow techniques for developers tuning how their coding assistants behave."}},{"@type":"Question","name":"Why do Cursor prompts work better when they specify exact file paths and functions instead of letting Cursor explore the codebase on its own?","acceptedAnswer":{"@type":"Answer","text":"Prescriptive prompts outperform open-ended exploration because Cursor works better without the obligation to guess. The most effective prompts named exact file paths, specified which functions to change, included examples of the canonical pattern, and added an explicit 'do not touch' section, turning Cursor into a constrained tool rather than a free-roaming assistant. Developers refining their Cursor prompting strategy can find similar workflow lessons on daily.dev."}}]} ```