Guides · Free tool · runs in your browser · no signup
CLAUDE.md generator
Answer a few questions about your project and get a clean, copy-paste CLAUDE.md for Claude Code — structured the way the current docs recommend. Everything runs locally; nothing you type is uploaded or stored.
Save it as CLAUDE.md at your repo root and commit it. Want it read by Cursor and Copilot too? Save the same content as AGENTS.md.
Want more than a starting point?
This gives you a solid, tailored CLAUDE.md. The AI Coding Workflow Pack ($24) goes further — 26 senior-level assets: ready-to-use subagents, slash commands, rules templates, and prompts for Claude Code, Cursor, and Copilot. Encode your standards once and stop re-explaining them.
What is a CLAUDE.md file?
CLAUDE.md is a plain-Markdown file that Claude Code reads at the start of every session in a project. It carries the context you'd otherwise repeat every time: the commands to run, where code lives, your conventions, and what not to touch. A project CLAUDE.md lives at your repository root and is checked into version control, so your whole team — and their agents — share the same instructions.
For a fuller walkthrough — the exact load order (managed → user → project → local), @imports, how it relates to AGENTS.md, and what to leave out — see the annotated CLAUDE.md example. Setting up reusable agents or commands next? The guide library covers subagents, skills, slash commands, hooks, and MCP.
Tips for a CLAUDE.md that actually helps
- Be specific and testable. "Run
pnpm testbefore every commit" beats "write good tests." The agent can act on the first. - Keep it short. Aim for under ~200 lines. It's loaded in full every session, so every line costs tokens and dilutes the important rules.
- Only durable facts. Put project-wide truths here; move path-specific guidance into
.claude/rules/and one-off notes intoCLAUDE.local.md(git-ignored). - Say what not to do. A short "Do not" list (generated files, migrations, direct commits to
main) prevents the most common mistakes.