When an AI coding agent like GitHub Copilot or Claude Code hits an ESLint error, it has two options: fix the code, or silence the linter. Fixing the code is harder. Silencing the linter is one comment.
AI agents will sometimes choose the easy path. They can insert any of these to make an error disappear:
// eslint-disable-next-line
Hi everyone,
TL;DR: Netlify now limits you to about 20 deploys per month. Every push to main costs credits. To push without using credits, add [skip netlify] to your commit message. Only leave it off when you want your live site to update.
What happened?
This project runs across multiple weeks in the same repo. Each week, your instructor delivers new instructions, checklists, and sometimes new tutorials or reference docs. These updates arrive as a GitHub Pull Request.
Your instructor creates a GitHub Issue in your repo titled something like "Move to Week 2." The issue describes what is changing and why — read it first.
A Pull Request appears linked to that issue. The PR contains updated files in the docs/ folder — new instructions, a new checklist, and any new tutorials for that week's concepts.
This guide walks you through a planning conversation with an AI agent. You should be reading this in your browser while your original "What Should I...?" repo is open in VS Code.
When you finish, you will have a BUILD-PROMPT.md file ready to take back to your my-advice repo.
Claude Code runs shell commands constantly — finding files, searching code, processing JSON, running scripts. The default Unix tools (find, grep, sed, node) work, but modern replacements written in Rust and Zig are significantly faster.
Installing these tools and telling Claude Code to prefer them can noticeably speed up your workflow.
| Default tool | Modern replacement | Speedup | Purpose |
|---|
The bare /compact command compresses your conversation but reads everything first—tool output, file contents, web fetches, search results. That takes minutes and often loses important context.
Create ~/.claude/commands/compact-guide.md with one line:
/compact Focus on: current task, file paths modified, decisions made, user preferences stated. Drop: tool output, search results, web content, file contents already acted on.
Builds on concepts from Introduction to Markdown.
Markdown files serve a dual purpose when working with AI coding assistants: they're human-readable documentation AND machine-parseable instructions. This tutorial explores the design considerations most users overlook.
When you write a README.md, you think about human readers. When you write a .github/copilot-instructions.md, you're writing for an AI that will interpret your instructions. But here's what many users miss: the AI reads your markdown differently than humans do.