archibate/dotfiles-claude

최근 기록된 소스 활동: 2026년 9월 26일 14:55

SkillsMP 카탈로그 업데이트: 2026년 11월 2일 18:02

수집된 skills: 27

GitHub 스타: 42
GitHub 포크: 11

AI 어시스턴트로 설치

이 프롬프트를 복사해 사용 중인 AI 어시스턴트에 입력하세요.

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/archibate/dotfiles-claude.

저장소 README

README 원본 파일: README.md

README 원본 보기

README 전체 펼치기

# dotfiles-claude Personal [Claude Code](https://docs.anthropic.com/en/docs/claude-code) configuration: settings, hooks, skills, and shell integrations. ## Install **Linux / macOS only.** Claude Code itself runs on Windows, but `setup.sh`, the integration shims, and every hook in this pack are bash. WSL works; native Windows shells are not supported. ```bash curl -fsSL https://raw.githubusercontent.com/archibate/dotfiles-claude/main/setup.sh | bash ``` > Prerequisites: `claude`, `git`, `jq`, `uv`, `node`, `npm`. The setup script is idempotent — re-run the same one-liner anytime to pull updates (it does `git pull --ff-only` on the existing checkout). ## Optional next steps After `setup.sh` finishes, a few commands cover the remaining one-time setup. Run whichever you want: ```bash # Wire the shell integration into your rc file (detects your shell, idempotent). bash ~/.claude/integration-install.sh # Tally the modern CLI tools preferred by CLAUDE.md against what's installed. claude "which CLI tools in ~/.claude/CLAUDE.md am I missing?" # Install kitty (if in a graphical session) and tmux if missing, then show best # practices for using Claude Code with them (clickable file:// links to delivered files via the # track-sent-file hook, peer-to-peer messaging via the claude-dm skill, # persistent sessions, etc.). claude "Check whether kitty (only if I'm in a graphical X/Wayland session) and tmux are installed. Install whichever is missing using my system package manager, then show best practices for using Claude Code with kitty and tmux. For reference, archibate's personal kitty and tmux configs live at https://github.com/archibate/dotfiles (kitty.conf) and https://github.com/archibate/dotfiles-tmux — fetch them and suggest cherry-picking what fits." ``` For reference, my personal configs that pair well with this setup: - **[archibate/dotfiles-tmux](https://github.com/archibate/dotfiles-tmux)** — tmux config tuned for the `claude-dm` peer-to-peer messaging skill and multi-session orchestration. - **[archibate/dotfiles](https://github.com/archibate/dotfiles)** — the rest (kitty, shell, neovim, etc.). Browse and cherry-pick what you want; there's no single one-liner installer — `kitty.conf`, `.zshrc`, etc. ## What's included - **settings.json** — permissions, hooks, MCP plugins, environment variables - **hooks/** — guardrails for safe tool use (block heredocs, enforce Write tool, etc.) - **skills/** — 53 skill packs (browser automation, translation, shader dev, and more) - **plugins/** — installed plugins via marketplaces (`claude-hud`, `claude-plugins-official`, `openai-codex`) - **integration.sh / .fish** — `claude` wrapper and model shortcuts - **bin/claude-commit** — standalone commit helper using Haiku - **integration-providers.sh / .fish** — optional shortcuts that route claude through third-party Anthropic-compatible endpoints (glm/deepseek/qwen/ofox; see the file header for the API key each maps to). - **CLAUDE.md** — global coding preferences and rules ## Defaults you should know **`permissions.defaultMode: "bypassPermissions"`** — every tool call runs without a Yes/No modal. The reasoning: prompts don't actually protect you. After a workday of clicking through them, your brain auto-accepts — you train the reflex without the judgment, and once you're burned out you're no different from `bypassPermissions`, except you spent your attention to get there. Either accept the risk and let claude work, or stay on web ChatGPT; the middle ground (prompts on, fingers on Enter) is the worst of both worlds. The actual safety layer lives in `hooks/no-*.sh` — mkfs / partition edits, writes to `/dev`, `/etc`, `/proc`, `/sys`, `/boot`, secure-delete, power-state, recursive chmod/chown, firewall flush, force-push, `git --amend`, `crontab -r`, `killall`, etc. All gates are soft reminders — there are explicit `# BYPASS_*_CHECK` markers so Claude can bypass one specific hazard to avoid trying bizarre workarounds. They are safety nets preventing accidental irreversible mistakes, in the belief that LLMs are not deliberately evil. Locking LLMs into a cage makes them do nothing but chat. To restore standard prompts: set `"defaultMode": "default"` in `settings.json`. ## Audit Hook An audit stop hook fires on Claude's final response after several edits, to review correctness and AI slop patterns, both in code and docs. It starts `claude` and `codex` headless; when issues are flagged, it reports to the main agent to ask it to fix them. > Bypass `codex` if not installed or not logged in. After used for a couple of weeks, you may show history audit stats: ```bash ~/.claude/hooks/audit-edits.py stats ``` Tweak `"env"` in `~/.claude/settings.json` to edit `"AUDIT_BACKEND": "both"` to `none|claude|codex|both` to configure.

저장소 README를 불러오는 중…

이 저장소의 skills

직업 카테고리 6개 · 63% 분류됨

skill

직업 분류

설명

업데이트

ast-grep

미분류

Translate natural language queries into ast-grep rules for structural code search. This skill should be used when a text-only grep would need 2+ alternations, regex capture groups, multi-line patterns, or filtering by syntactic role (call vs. import vs.…

원문 언어: 영어

2026-09-26

babysit

미분류

Run long-running tasks under babysit — supervised background runner with cgroup-enforced memory/CPU caps and observability-stall kill. Use BEFORE any long-running task (>2 min), compute-intensive jobs (e.g. builds/benchmarks/renders/training), background…

원문 언어: 영어

2026-09-26

better-translate

미분류

Translate English into Chinese in natural human voice. Use before writing or polishing any Chinese translation of English content.

원문 언어: 영어

2026-09-26

employee-chat

미분류

用自然、简洁、有判断力的员工口吻撰写或改写工作聊天消息。在向上级汇报、同事协作、客户沟通，以及用户要求工作消息少些 AI 腔、像真人员工说话时使用。

원문 언어: 중국어

2026-09-26

grill-me

미분류

Interview the user about a plan or design until reaching shared understanding. This skill should be used when the user wants to stress-test a plan, get grilled on their design, or mentions "interview me".

원문 언어: 영어

2026-09-26

mini-prompt

미분류

Form a minimal self-contained prompt to reproduce this session. Use when user want to form a minimal spec of existing system, or when they migrate from a degraded session to a fresh one.

원문 언어: 영어

2026-09-26

minimalist

미분류

Use minimalist language.

원문 언어: 영어

2026-09-26

cpp-hpc-optimization

미분류

Evidence-driven C++ high-performance computing design, profiling, and optimization across algorithmic time/space complexity, data layout, cache and memory behavior, numerics, SIMD, multicore scheduling, and CPU/accelerator pipelines. Use before designing or…

원문 언어: 영어

2026-09-25

cpp-oop-style

미분류

High-quality C++ OOP coding style (archibate / parallel101 lineage) that overrides sloppy AI-default C++. Use this skill WHENEVER writing, editing, refactoring, or reviewing C++ code (.cpp / .h / .hpp / .cc / .cxx), designing C++ classes, interfaces, APIs, or…

원문 언어: 영어

2026-09-25

skill-creator

미분류

Create, edit, and benchmark agent skills — including evals with variance analysis and optimizing a skill's description for triggering accuracy.

원문 언어: 영어

2026-09-25

writing-prompt

기타 컴퓨터 관련 직업

Create or edit agent-facing/LLM prompts using modern prompt-engineering practices. Use this skill before editing agent-facing docs, rule files, references, skills, memory, or any form of LLM prompt. Also use it before writing LLM tests or evaluations. This is…

원문 언어: 영어

2026-09-13

read-url

소프트웨어 개발자

Extract clean, complete markdown from any web page — articles, docs, READMEs, blog/social posts, academic papers. Also use as a fallback when curl returns noisy HTML or WebFetch returns truncated, summarized, or refused results.

원문 언어: 영어

2026-09-12

context7

소프트웨어 개발자

Fetch current docs and code examples for third-party libraries, SDKs, APIs, and CLI tools via Context7. Use before writing or editing non-stdlib library calls — your prior knowledge may be stale. Prefer over web search for library docs.

원문 언어: 영어

2026-08-11

grep-app

소프트웨어 개발자

Code search across millions of public GitHub repositories via grep.app. Use before writing or planning code to find real-world usage of libraries, frameworks, APIs, algorithms, or syntax patterns. Prefer this over web search for code examples.

원문 언어: 영어

2026-08-11

fresh-arch

소프트웨어 개발자

Architecture design reasoned forward from requirements. Use when the user wants a fresh-mind design pass — "design an architecture for", "how would you build", "redesign from scratch" — or when repeated follow-up patches accumulate on the same module.…

원문 언어: 영어

2026-08-10

claude-dm

소프트웨어 개발자

Send peer-to-peer message and control to Claude Code sessions. Use when coordinating multiple simultaneous Claude sessions, or user says "dm another claude", "list claude sessions", "peek peer claude", "spawn claude in tmux", "monitor remote claude", "notify…

원문 언어: 영어

2026-07-23

zhlint

편집자

中文排版规范（全角标点、中英文空格）+ zhlint 校验工具。Use whenever writing or editing formal Chinese document content — Chinese markdown docs, README 中文版, blog posts, WeChat articles, translation output, announcements, tutorials — or when the user mentions 排版, 全角/半角, punctuation…

원문 언어: 중국어

2026-07-20

memory-add

기타 컴퓨터 관련 직업

Append a single bullet to staging memory. Use when a durable fact, knowledge, lesson, or pitfall (mistake-correction pattern) emerged mid-conversation that's potentially worth persisting into long-term memory. Also use when user corrected your mistake, says…

원문 언어: 영어

2026-06-23

claude-headless

소프트웨어 개발자

Programmatic/headless Claude Code usage — CLI flags, stream-json protocol, Agent SDKs, session management. Use before spawning Claude as a subprocess, testing skill or agent behavior, running "claude -p", or building with the Agent SDK.

원문 언어: 영어

2026-06-05

agent-browser

소프트웨어 개발자

Inspect or debug web pages in a headless browser — fill forms, click buttons, take screenshots, test web apps, review frontend UI/UX aesthetics. Use for headless browser automation. This skill should be used when user says "test web app", "headless browser",…

원문 언어: 영어

2026-05-29

note

비서 및 행정 보조원(법률, 의료 및 임원 제외)

Add a note only visible to user. A private scratchpad not visible in agent context.

원문 언어: 영어

2026-05-20

onesent

기술 작가

One sentence output style

원문 언어: 영어

2026-05-19

chrome-cdp

소프트웨어 개발자

Connect to user's local Chrome session. Inspect and interact with web pages in a headful browser shared with user cookies and login states — navigate pages, click buttons, take screenshots, test web apps in headful Chrome the user could see in real-time. Use…

원문 언어: 영어

2026-05-14

tmux

네트워크·컴퓨터 시스템 관리자

Spawn and interact with interactive CLIs (python, gdb, etc.) in tmux — send keystrokes, scrape pane output. This skill should be used when running any interactive CLI command, starting program requires interactive stdin / PTY / TUI render, or when the user…

원문 언어: 영어

2026-05-14

jina-ai

소프트웨어 개발자

Web search with time-window, region/language, and `site:` filters, academic papers (arXiv/SSRN), PDF table/figure extraction, BibTeX, image search, web page reading, embeddings, reranking, classification, and deduplication (text or images) via Jina AI. Use…

원문 언어: 영어

2026-05-11

repo-cache

소프트웨어 개발자

Clone or reuse a cached local checkout of a remote git repo (GitHub/GitLab/Bitbucket). Use when the user gives a repo URL or `owner/repo` identifier (e.g. "facebook/react"), says "clone this repo" / "read the source", or needs to browse a remote codebase.

원문 언어: 영어

2026-05-06

scrapling

소프트웨어 개발자

Scrape, crawl, or extract data from websites using Scrapling — anti-bot bypass (Cloudflare Turnstile), stealth headless, spiders, adaptive scraping, JS rendering. Also a fallback when WebFetch fails due to anti-bot protections or JS rendering.

원문 언어: 영어

2026-05-06

수집된 skill 27개 중 27개를 표시합니다.