micky@garden:~/claude-code-artifact-lab$cd ..

Claude Code courseware

Claude Code Artifact Lab

A focused workbench for drafting Claude memory, skills, and subagents before learners use them with Claude Code.

3artifact types4pathways0issues
CLAUDE.mdrepo root or .claude/CLAUDE.md
Validation — no blocking issues

Project Starter

CLAUDE.md - Project Starter

3,401 chars

Live preview

CLAUDE.md

CLAUDE.md

Claude Code Notes

  • Claude Code reads CLAUDE.md at session start; keep this file concise, specific, and under 200 lines.
  • Use this file for project facts Claude should know every session: commands, conventions, architecture, and repeat corrections.
  • Move multi-step procedures into .claude/skills/<skill-name>/SKILL.md.
  • Move path-specific conventions into .claude/rules/*.md with paths frontmatter when the rule only applies to part of the repo.

Fill This In First

MaterialStarter value
Memory location./CLAUDE.md for root project memory, or ./.claude/CLAUDE.md when the team keeps Claude config together
Claude Code behaviorPlan mode guidance, memory boundaries, rules, skills, permissions, or verification conventions
First files to inspectREADME files, package manifests, CI config, docs, and existing scripts
Common Claude Code taskinstall dependencies, run the right checks, and make a small safe change
Main risktoo much generic guidance hides the few project facts Claude should load every session

Memory Boundaries

  • Keep always-loaded memory short: commands, architecture landmarks, repeat corrections, and hard constraints.
  • Every line should be something Claude needs in most sessions; remove tutorials, stale notes, and generic advice.
  • Put path-specific rules in .claude/rules/ so they load only when matching files are relevant.
  • Put repeatable procedures in skills so their full instructions load only when invoked.

Commands

  • Install dependencies: pnpm install
  • Lint: pnpm run lint
  • Typecheck: pnpm run typecheck
  • Test: pnpm run test
  • Build: pnpm run build

Project Context

  • Context source to inspect first: README files, package manifests, CI config, docs, and existing scripts.
  • Primary audience: Claude Code sessions entering a shared project.
  • Common Claude Code task: install dependencies, run the right checks, and make a small safe change.

Verification

  • Prefer focused checks that match the edit before running the full suite.
  • For UI changes, capture a browser screenshot or describe the visual check performed.
  • Report commands run and any checks skipped with the reason.
  • Use /memory to inspect loaded project memory when behavior does not match this file.

Rules And Skills Split

  • Use .claude/rules/testing.md for test conventions that only matter near test files.
  • Use .claude/rules/frontend.md for UI conventions that only apply under frontend paths.
  • Use .claude/skills/<skill-name>/SKILL.md for repeatable workflows with inputs, steps, and outputs.
  • Keep this file as navigation and durable memory, not a long procedure manual.

Critical Warnings

  • Do not store personal sandbox URLs, private tokens, or local-only preferences here; put them in CLAUDE.local.md and gitignore it.
  • Remove outdated or contradictory rules during review.
  • Risk to watch for this pathway: too much generic guidance hides the few project facts Claude should load every session.

Local Notes

  • Personal notes belong in CLAUDE.local.md, not in this shared file.
  • Personal cross-project preferences belong in ~/.claude/CLAUDE.md.

Test Prompt

Ask Claude Code: "Summarize the CLAUDE.md, CLAUDE.local.md, and .claude/rules files you loaded for this repo. What would you do before a Claude Code project onboarding change?"

Claude Code project onboardinginstall dependencies, run the right checks, and make a small safe change

template guide

What this artifact should include.

Use these notes as source-backed guidance for the selected artifact. The editor only flags clear errors and warnings; this section stays advisory.

Starter materials

  • Project Starter source pack

    For Claude Code project onboarding, gather README files, package manifests, CI config, docs, and existing scripts.

    Use it to help Claude Code sessions entering a shared project install dependencies, run the right checks, and make a small safe change.
  • Project memory scope

    Choose whether this shared project memory belongs at `./CLAUDE.md` or `./.claude/CLAUDE.md`, then keep it focused on facts Claude should know every session.

    Commands, verification expectations, architecture landmarks, repeated corrections, and critical warnings.
  • Every-session memory

    Keep only the facts Claude should load every session: commands, architecture landmarks, repeat corrections, and hard constraints.

    Under 200 lines, concrete commands, no generic advice like 'write clean code'.
  • Rules and skills split

    Move path-specific conventions to `.claude/rules/` and repeatable procedures to `.claude/skills/` instead of bloating CLAUDE.md.

    Frontend styling rules in `.claude/rules/frontend.md`; release workflow in `.claude/skills/release/SKILL.md`.
  • Local/private notes

    Separate personal sandbox URLs, machine-specific shortcuts, and private preferences into CLAUDE.local.md or user-level memory.

    Add `CLAUDE.local.md` to `.gitignore`; do not commit private tokens or local-only paths.
  • Load test

    Keep a prompt that asks Claude Code to summarize which memory files loaded and which instructions apply to a sample edit.

    "Which CLAUDE.md, CLAUDE.local.md, and .claude/rules files shaped your plan for this change?"

Guidelines

  • Starts with a clear H1
  • Names the work context
  • Includes a way to test the artifact
  • Documents concrete commands
  • Keeps Claude Code memory focused
  • Keeps memory concise and concrete
  • Names verification expectations
  • Separates personal local notes

Examples

Good

Good: source-backed Claude Code project onboarding guidance with a clear trigger, scoped commands or tools, concrete verification, and no duplicated generic advice.

Needs work

Bad: generic advice such as "be helpful", broad tool grants, no trigger, no source boundaries, and no way to verify the result.

References

  • Claude Code memory docs

    Explains CLAUDE.md scope, concise instruction writing, load order, local memory, and rules.

  • Claude Code best practices

    Recommends concise project memory, clear verification, skills for reusable workflows, and subagents for focused investigation.

  • Explore the .claude directory

    Maps where Claude Code reads project memory, rules, skills, subagents, hooks, settings, and local-only files.

Test prompt

Use this artifact with a short Claude Code project onboarding sample. Ask the agent to install dependencies, run the right checks, and make a small safe change, then report which instruction or reference shaped the result.