CLAUDE.md · luongnv89/claude-howto
CLAUDE.md from luongnv89/claude-howto: 63 lines covering Critical commands, Architecture map, Hard rules, Workflow preferences and more.
Template by luongnv89
Install
Works with Claude Code. Read the repository before you give any tool access to your data.
curl -fsSL https://raw.githubusercontent.com/luongnv89/claude-howto/main/CLAUDE.md -o CLAUDE.md
Read it and adapt it to your project before use.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: prompt-engineering
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
What's in it
A CLAUDE.md file: project memory Claude Code loads at the start of every session.
- For
- Claude Code
- Length
- 63 lines
- File
CLAUDE.md
Sections
- Critical commands
- Architecture map
- Hard rules
- Workflow preferences
- Token Efficiency
First lines (shown under the repo's MIT license)
# CLAUDE.md Tutorial repo. Output is markdown in numbered modules `01-` through `10-`, not an app. Scripts in `scripts/` exist only to validate docs and build the EPUB. See also [CONTRIBUTING.md](CONTRIBUTING.md#3-set-up-your-environment) for stack and setup (uv, Node.js >= 20, pre-commit) and [STYLE_GUIDE.md](STYLE_GUIDE.md) for lesson structure. ## If the user is learning Many people clone this repo to learn, not to contribute. Unless the user asks for a change to the repo, treat the rules below as maintainer conventions, not their preferences: don't edit lessons, propose repo fixes, or set up hooks for them. Point them to a `CLAUDE.local.md` they create at the repo root (gitignored, loaded alongside this file) for their own context. When they do ask for a change, commit, or PR, every rule below applies in full. ## Critical commands ```bash # Quality gate (also runs on commit via pre-commit hooks) pre-commit run --all-files # Tests pytest scripts/tests/ -v # EPUB build (renders Mermaid with Pele — needs Node.js >= 20; the pinned bundle downloads once, then works offline) uv run scripts/build_epub.py # Python tooling ruff check scripts/ && ruff format scripts/ mypy scripts/ --ignore-missing-imports bandit -c scripts/pyproject.toml -r scripts/ --exclude scripts/tests/ ``` Pre-commit runs 5 doc checks on `.md` changes: markdown-lint, cross-references, mermaid-syntax, link-check, markdown-rendering. All must pass. The EPUB build is **not** a pre-commit hook — it runs in CI only (`build-epub` job in `.github/workflows/test.yml`), because it is a slow full-book build; the mermaid-syntax hook already renders every diagram with the same Pele bundle. ## Architecture map - `01-` … `10-` — tutorial modules. **Numbered prefix = learning order**, not alphabetical. Do not reorganize. - Each module: `README.md` + copy-paste templates (`.md`, `.json`, `.sh`). - `scripts/` — utilities (EPUB builder, link/mermaid/cross-ref validators). Not the product. - `02-memory/*.md` — CLAUDE.md templates users copy into their own projects. Don't confuse with this file. ## Hard rules - **YOU MUST NOT commit or push without explicit user request.** - **YOU MUST NOT add `Co-Authored-By: Claude`** to any commit message. - Always activate `.venv` before running Python scripts (check `venv/`, `.venv/`, `env/`). - Internal links use **relative paths** (e.g. `01-slash-commands/README.md`); anchors use `#heading-name`. - Code fences **must** declare a language (`bash`, `python`, `json`, …) — the cross-reference check fails otherwise. - External URLs must be reachable and stable. No ephemeral links. - Mermaid diagrams must render with Pele. The pre-commit mermaid hooks get Node.js 22 from pre-commit and run strict, so they fail instead of skipping; the first run needs network to download Node.js and the pinned Pele bundle (offline: `SKIP=mermaid,vietnamese-mermaid,japanese-mermaid`). Run by hand, `scripts/check_mermaid.py` only warns without Node.js >= 20, and fails under `--strict` or in CI. A broken EPUB build is usually invalid Mermaid, a missing Node.js >= 20, or a failed bundle download. - Commit format: `type(scope): subject` where `scope` matches the module folder (e.g. `feat(slash-commands):`, `docs(memory):`, `fix(README):`). - Do not reorganize the `01-`–`10-` numbering. The order is the curriculum. ## Workflow preferences - For lesson edits, follow `STYLE_GUIDE.md` for structure/naming/diagrams. - Small fixes → minimal diff. Don't rewrite a section to fix a typo. - When adding a module page: README + templates first, then update root `README.md` index and `LEARNING-ROADMAP.md` if order/timing changes. - Tutorial > library: prioritize clear explanations and copy-paste examples over reusable abstractions. - If a quality check fails, fix the underlying issue. Don't bypass with `--no-verify`. ## Token Efficiency - Never re-read files you just wrote or edited. You know the contents. - Never re-run commands to "verify" unless the outcome was uncertain. - Don't echo back large blocks of code or file contents unless asked. - Batch related edits into single operations. Don't make 5 edits when 1 handles it. - Skip confirmations like "I'll continue..." Just do it. - If a task needs 1 tool call, don't use 3. Plan before acting. - Do not summarize what you just did unless the result is ambiguous or you need additional input.
Signals
- Stars
- 44,359
- Forks
- 5,482
- Open issues
- 15
- Last commit (this folder)
- Oct 19, 2026 (9 days ago)
- Last release
- Jun 2, 2026 (4 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/luongnv89/claude-howto
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from luongnv89
- 0.824score
skill-index-updater
luongnv89
Add GitHub skill repos to the ASM index: clone, audit, eval, regenerate index, rebuild catalog, open PR.
SkillAFreshness: A★ 959Claude Code - 0.824score
refresh-index
luongnv89
Sync every enabled repo in the curated skill index and open a confirmation-gated PR.
SkillAFreshness: A★ 959Claude Code +9 - 0.821score
skill-install-improved
luongnv89
Install an improved variant of one named skill: resolve it by local path, repo, or name, run skill-creator's retrofit on a throwaway copy, then install the…
SkillAFreshness: A★ 959Claude Code +9 - 0.821score
skill-upstream-pr
luongnv89
Improve an open-source GitHub skill and open a friendly suggestion PR upstream: fork, run skill-creator's retrofit, attach asm eval before/after metrics.
SkillAFreshness: A★ 959Claude Code +9 - 0.821score
skill-shortener
luongnv89
Refactor a too-long SKILL.md by progressive disclosure: measure token cost, classify every section KEEP/CUT/MOVE, shorten the body into references/ and…
SkillAFreshness: A★ 959Claude Code +9 - 0.807score
well-formed
luongnv89
Review pull request diffs for code smells, style issues, and safety problems before merging.
SkillAFreshness: A★ 959Claude Code +9
Cite this page
CLAUDE.md · luongnv89/claude-howto. Agent Mart. https://agentmart.dev/templates/luongnv89-claude-howto-claude-md (last updated 2026-10-24)