<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/i-rebuilt-my-claude-code-workflow-around-its-creator-s-advice-and-the-difference-surprised-me-8hnbqrnow" -->

---
title: I rebuilt my Claude Code workflow around its creator's...
description: Claude Code creator Boris Cherny has reversed some of his own past advice on using the tool effectively. He now skips Plan Mode for synchronous CLI work,...
canonical: https://daily.dev/posts/i-rebuilt-my-claude-code-workflow-around-its-creator-s-advice-and-the-difference-surprised-me-8hnbqrnow
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: I rebuilt my Claude Code workflow around its creator's advice, and the difference surprised me | daily.dev
og:description: Claude Code creator Boris Cherny has reversed some of his own past advice on using the tool effectively. He now skips Plan Mode for synchronous CLI work,...
og:url: https://daily.dev/posts/i-rebuilt-my-claude-code-workflow-around-its-creator-s-advice-and-the-difference-surprised-me-8hnbqrnow
og:image: https://api.daily.dev/og/posts/8hnbqrNOw.png
og:image:alt: I rebuilt my Claude Code workflow around its creator's advice, and the difference surprised me
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 rebuilt my Claude Code workflow around its creator's advice, and the difference surprised me

**[XDA Developers](https://daily.dev/sources/xda-developers)** · 9 min read · 1 upvotes · 0 comments

## Summary

Claude Code creator Boris Cherny has reversed some of his own past advice on using the tool effectively. He now skips Plan Mode for synchronous CLI work, favoring Auto Mode, which uses a security classifier to approve actions instead of manual confirmation. He recommends documenting Claude's mistakes in CLAUDE.md or a skill file so it doesn't repeat them, and having Claude actually test what it builds, including GUI interaction via computer use. He also now favors minimal system prompts and tools over heavy context, and suggests using Anthropic's Agent SDK to build loops that trigger agents automatically on events like new GitHub issues.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.xda-developers.com/rebuilt-my-claude-code-workflow-around-its-creators-advice>

## Questions this post answers

### Should I still use Plan Mode in Claude Code or has that advice changed?

Claude Code creator Boris Cherny no longer uses Plan Mode for his synchronous CLI workflow. He says newer models are capable enough to skip a separate planning phase, and constant approval prompts cause permission fatigue where people start accepting actions without reading them. He now favors Auto Mode instead, which uses a security classifier to evaluate tool actions automatically.

_Following how Claude Code best practices shift helps you avoid relying on stale workflow advice; track updates like these on daily.dev._

### How can I stop Claude Code from repeating the same coding mistakes?

Document each mistake and what was learned from it directly in the CLAUDE.md file or a dedicated skill file, as Boris Cherny recommends, rather than just correcting Claude in chat and moving on. That gives Claude a persistent record to reference in future sessions so it doesn't repeat the same error again.

_Developers refining AI coding workflows can keep up with techniques like this on daily.dev._

### What is Anthropic's /goal command used for in Claude Code?

The /goal command lets a developer define a specific outcome and have Claude continue iterating on its own until it meets the stated criteria or hits a set limit, rather than requiring repeated manual correction. It pairs with goal-based testing loops, such as having Claude launch and interact with a GUI application to verify its own work before resubmitting changes.

_Anyone building autonomous coding loops can track new Claude Code features like this on daily.dev._

## Similar posts on daily.dev

- [Claude Code felt overhyped until I discovered the workflow that makes it genuinely powerful](https://daily.dev/posts/claude-code-felt-overhyped-until-i-discovered-the-workflow-that-makes-it-genuinely-powerful-ellcs8hij) · XDA Developers · 0 upvotes · 0 comments
- [Inside the Development Workflow of Claude Code's Creator](https://daily.dev/posts/inside-the-development-workflow-of-claude-code-s-creator-rmtwa51nd) · InfoQ · 2 upvotes · 1 comments
- [Claude Code Tips I Wish I’d Had From Day One](https://daily.dev/posts/claude-code-tips-i-wish-i-d-had-from-day-one-0mqrs17ox) · marmelab · 5 upvotes · 0 comments
- [Claude Code — A Practical Guide to Automating Your Development Workflow](https://daily.dev/posts/claude-code-a-practical-guide-to-automating-your-development-workflow-pbqztpxro) · We Are .NET · 0 upvotes · 0 comments

---

Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#anthropic](https://daily.dev/tags/anthropic), [#claude-code](https://daily.dev/tags/claude-code), [#prompt-engineering](https://daily.dev/tags/prompt-engineering), [#vibe-coding](https://daily.dev/tags/vibe-coding)

[View this post on daily.dev](https://daily.dev/posts/i-rebuilt-my-claude-code-workflow-around-its-creator-s-advice-and-the-difference-surprised-me-8hnbqrnow)

```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 rebuilt my Claude Code workflow around its creator's advice, and the difference surprised me","url":"https://daily.dev/posts/i-rebuilt-my-claude-code-workflow-around-its-creator-s-advice-and-the-difference-surprised-me-8hnbqrnow","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/i-rebuilt-my-claude-code-workflow-around-its-creator-s-advice-and-the-difference-surprised-me-8hnbqrnow"},"datePublished":"2026-10-23T16:33:39.849Z","dateModified":"2026-10-23T16:36:34.376Z","description":"Claude Code creator Boris Cherny has reversed some of his own past advice on using the tool effectively. He now skips Plan Mode for synchronous CLI work,...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e38ffed5b6f227b0f9ec5f12379c1414?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e38ffed5b6f227b0f9ec5f12379c1414?_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-rebuilt-my-claude-code-workflow-around-its-creator-s-advice-and-the-difference-surprised-me-8hnbqrnow","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,anthropic,claude-code,prompt-engineering,vibe-coding","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 rebuilt my Claude Code workflow around its creator's advice, and the difference surprised me"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/i-rebuilt-my-claude-code-workflow-around-its-creator-s-advice-and-the-difference-surprised-me-8hnbqrnow#faq","mainEntity":[{"@type":"Question","name":"Should I still use Plan Mode in Claude Code or has that advice changed?","acceptedAnswer":{"@type":"Answer","text":"Claude Code creator Boris Cherny no longer uses Plan Mode for his synchronous CLI workflow. He says newer models are capable enough to skip a separate planning phase, and constant approval prompts cause permission fatigue where people start accepting actions without reading them. He now favors Auto Mode instead, which uses a security classifier to evaluate tool actions automatically. Following how Claude Code best practices shift helps you avoid relying on stale workflow advice; track updates like these on daily.dev."}},{"@type":"Question","name":"How can I stop Claude Code from repeating the same coding mistakes?","acceptedAnswer":{"@type":"Answer","text":"Document each mistake and what was learned from it directly in the CLAUDE.md file or a dedicated skill file, as Boris Cherny recommends, rather than just correcting Claude in chat and moving on. That gives Claude a persistent record to reference in future sessions so it doesn't repeat the same error again. Developers refining AI coding workflows can keep up with techniques like this on daily.dev."}},{"@type":"Question","name":"What is Anthropic's /goal command used for in Claude Code?","acceptedAnswer":{"@type":"Answer","text":"The /goal command lets a developer define a specific outcome and have Claude continue iterating on its own until it meets the stated criteria or hits a set limit, rather than requiring repeated manual correction. It pairs with goal-based testing loops, such as having Claude launch and interact with a GUI application to verify its own work before resubmitting changes. Anyone building autonomous coding loops can track new Claude Code features like this on daily.dev."}}]}
```

