Summarize your architecture rules and give your AI a persistent memory Summarize your architecture rules and give your AI a persistent memory TL;DR: When you use the /init command, you create a context file that saves you from repeating instructions in every new session. TL;DR: When you use the /init command, you create a context file that saves you from repeating instructions in every new session. Common Mistake ❌ You waste time copying and pasting the same project rules into every new chat. You forgot to tell the AI about your specific commands and instructions. When you start a fresh session, the AI loses all previous context about your coding style or architecture. Problems Addressed 😔 Repetitive and manual meta-prompting
Inconsistent code styles across sessions
High token usage from redundant instructions
AI hallucinations about build or test commands
Onboarding new projects Repetitive and manual meta-prompting Inconsistent code styles across sessions code styles High token usage from redundant instructions AI hallucinations about build or test commands Onboarding new projects How to Do It 🛠️ Open your terminal in your project root directory.
Type /init if you use Claude Code or OpenCode.
Review the generated CLAUDE.md or AGENTS.md file.
Add specific "landmines" or things the AI can't see in the code.
Save the file, review the commit diffs, and start your next coding task.
You can do it again when you have new rules.
Create your own system prompts, persist them, and keep an eye on them regularly. Open your terminal in your project root directory. Type /init if you use Claude Code or OpenCode. /init Review the generated CLAUDE.md or AGENTS.md file. CLAUDE.md AGENTS.md Add specific "landmines" or things the AI can't see in the code. Save the file, review the commit diffs, and start your next coding task. review the commit diffs You can do it again when you have new rules. Create your own system prompts, persist them, and keep an eye on them regularly. system prompts Benefits 🎯 When you initialize (/Init) your project, the AI reads all the repository immediately. /Init You get consistent results without extra effort. You save tokens because you don't have to describe your stack every time. Context 🧠 AI assistants start every session with a blank slate. They don't remember what you told them yesterday. Tools like Claude Code look for a specific file to understand your project. When you provide this saved file, you close the gap between separate AI sessions. Prompt Reference 📝 Bad Prompt: Bad Prompt: Remember to use 2 spaces for indentation 
and always run 'npm run test:unit' 
before you finish the task. Remember to use 2 spaces for indentation 
and always run 'npm run test:unit' 
before you finish the task. Good Prompt: Good Prompt: git clone https://github.com/torvalds/linux.git

/Init git clone https://github.com/torvalds/linux.git

/Init Considerations ⚠️ Keep the generated file short. When you add too much text, the AI might get confused. get confused Don't include information that the AI can already find in your package.json or README.md. package.json README.md Type 📝 [X] Automatic Limitations ⚠️ You can only use this tip with AI tools that support persistent context files. If you change your build tools, you can update the Markdown file manually, run /init again, or do both. In Windsurf, there is no literal /init terminal command that automatically generates a configuration file, as Claude Code or OpenCode does. Windsurf uses a more agentic approach where the AI (Cascade) either discovers your rules automatically or creates them when you ask. Try this prompt: Cascade: initialize a .windsurfrules file for this project. Include my tech stack, coding standards, and build commands. Cascade: initialize a .windsurfrules file for this project. Include my tech stack, coding standards, and build commands. Tags 🏷️ Standards Standards Level 🔋 [X] Beginner Related Tips 🔗 https://hackernoon.com/ai-coding-tip-006-review-every-line-before-commit?embedable=true https://hackernoon.com/ai-coding-tip-006-review-every-line-before-commit?embedable=true Use .cursorrules for local preferences
Create modular AGENTS.md files for monorepos
Keep your README.md updated for better AI indexing Use .cursorrules for local preferences .cursorrules Create modular AGENTS.md files for monorepos AGENTS.md Keep your README.md updated for better AI indexing README.md Conclusion 🏁 When you use /init, you turn your AI from a temporary contributor into a permanent team member. /init More Information ℹ️ https://opencode.ai/docs/rules/?embedable=true https://opencode.ai/docs/rules/?embedable=true https://docs.anthropic.com/en/docs/agents-and-tools/claude-code/overview%23claudemd-guide?embedable=true https://docs.anthropic.com/en/docs/agents-and-tools/claude-code/overview%23claudemd-guide?embedable=true https://docs.cursor.com/context/rules-for-ai?embedable=true https://docs.cursor.com/context/rules-for-ai?embedable=true https://docs.windsurf.com/windsurf/cascade/memories?embedable=true https://docs.windsurf.com/windsurf/cascade/memories?embedable=true Also Known As 🎭 Project-AI-Onboarding
Context-Bootstrapping Project-AI-Onboarding Context-Bootstrapping Tools 🧰 Claude Code
OpenCode
Cursor (via .cursorrules) Claude Code OpenCode Cursor (via .cursorrules) Disclaimer 📢 The views expressed here are my own. I am a human who writes as best as possible for other humans. I use AI proofreading tools to improve some texts. I welcome constructive criticism and dialogue. I shape these insights through 30 years in the software industry, 25 years of teaching, and writing over 500 articles and a book. This article is part of the AI Coding Tip series. AI Coding Tip

This story contains AI-generated text. The author has used AI either for research, to generate outlines, or write the text itself. 

Keep

AI Coding Tip 010 - Access All Your Code

Buy My Book

Share your ideas with me!

AI Coding Tip 011 - How to Initialize Agents.md

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

10 Predictions About the Future of Finance

AI-Assisted Coding with Tabnine

DOE v. Github: Code Stored on GitHub Ultimately Belong to Their Authors

DOE v. Github: Defendants' Conduct Raises Questions Related to DMCA Violations, Privacy, Data Breach

DOE v. Github: Defendants Never Sought Permission to Use Others' Code to Train Codex or Copilot

DOE v. Github: Lawsuit Falls Under California's Jurisdiction Because of GitHub & OpenAI's HQ

10 Predictions About the Future of Finance

AI-Assisted Coding with Tabnine

DOE v. Github: Code Stored on GitHub Ultimately Belong to Their Authors

DOE v. Github: Defendants' Conduct Raises Questions Related to DMCA Violations, Privacy, Data Breach

DOE v. Github: Defendants Never Sought Permission to Use Others' Code to Train Codex or Copilot

DOE v. Github: Lawsuit Falls Under California's Jurisdiction Because of GitHub & OpenAI's HQ

Light-Mode

Classic

Newspaper

Minty

Dark-Mode

Neon Noir

Minty

HN StartUps