REA
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
This repo is a collection of 29 skills
Agent skills that make DeepSeek models plan, verify and ship like frontier models. npx skills add aicodeking/deepseek-stack作者の方は、このコードを README に貼れます。バッジはこのページにリンクします。
[](https://mcp.popnx.com/en/skills/aicodeking-deepseek-stack)同じ作業で AI をサービスにつなぐ MCP
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
Codebase intelligence for TypeScript and JavaScript, exposed as structured MCP analysis tools.
MCP server for terminal commands, file operations, and process management
同じ種類の作業。Skill 一覧と同じ並び
| これ deepseek-stack | rea | andrej-karpathy-skills | graphify | MoneyPrinterTurbo | |
|---|---|---|---|---|---|
| 作者 | aicodeking | morluto | multica-ai | Graphify-Labs | harry0703 |
| Skill 数 | 29 | 1 | 1 | 1 | 1 |
| GitHub スター | 299 | 463,670 | 233,682 | 148,887 | 143,987 |
| 最終コード更新 | 2026年10月29日 | 2026年11月17日 | 2026年4月20日 | 2026年11月17日 | 2026年11月16日 |
| ライセンス | MIT | MIT | – | Apache-2.0 | MIT |
aicodeking による 29 個の Skill(指示セット)。AI が読んでそのとおりに作業します · MCP server ではありません
Install only the ones you use. Open a skill to see how to install it
If you want every one, let the AI install them in one go. Know 2 things first
Uploading to Claude on the web and apps: you need 29 uploads, one ZIP per skill. Open each skill in the list below to download it
Checked against official docs 2026年11月12日 · Source: support.claude.com, code.claude.com, learn.chatgpt.com, cursor.com
Takes over an existing PR, branch or agent thread and finishes its original goal while keeping credit for the first contributors. Use when the user says take over, adopt, pick up or finish this PR, or /ds-adopt.
Run N independent attempts at the same task in parallel, pick the strongest as the base, and graft the best parts of the others into it. Use when one attempt at a non-trivial artifact could land on the wrong shape, or for /ds-best-of-n, 'try a few versions', 'best of 3'.
Find what a change could break outside its own diff, then prove the key safety fact by running real code. Use before shipping a risky or untrusted change, for 'what could this break', 'blast radius of X', or /ds-blast-radius.
Design the types, function signatures and module layout first, agree on that shape, then implement against it. Use when work is non-trivial and coding straight away would lock in the wrong structure, or for /ds-blueprint, 'design this first', 'architect this'.
Give the user a short plain-language catch-up on what changed since their last message, followed by every decision still waiting on them. Use when the user returns after a long or unattended run, asks 'where are we', 'what did I miss', or types /ds-catchup.
Keep a PROGRESS.md for long tasks and move to a fresh session or subagent that reads only that file when the context gets long or stale. Use when a task spans many steps or sessions, when the model starts repeating itself, forgets earlier facts, or re-tries old fixes, or when the context window is…
Builds a frontend with a clear visual direction instead of the generic AI look. Use when you create a new page, app screen, or component, or restyle one that looks templated. Run as /ds-design.
Reviews an existing UI with scores based on evidence, then fixes the highest-impact problems. Use when a page or app screen looks generic, cluttered, or unfinished, or before you ship UI work. Run as /ds-design-audit.
Explain how a piece of code works, why it is built that way, and what that means, in plain language a person can follow. Use for 'how does X work', 'why is it like this', 'teach me this', onboarding to a subsystem, or /ds-explain.
Pick the right deepseek-stack skill for a task and explain how the pack fits together. Use when the user asks which skill to use, what the pack contains, or types /ds-help.
Gets independent reviews of a change from two or three different models or providers and loops on fixes until they agree. Use when the user says jury, panel review, second opinions, ask other models, or /ds-jury.
Keep a one-row-per-decision log during long or unattended runs, and re-read it before each step so the run does not drift from its goal. Use for multi-phase or autonomous work, work a human will review later, or /ds-ledger.
Find mistakes agents keep repeating in this repo and turn each into structure that blocks it, preferring types over lint over tests over docs, plus an end-of-session reflection. Use after a user correction, when the same fix comes up twice, at the end of a long session, or for /ds-lessons.
Always-on working style for DeepSeek agents: restate the task, read before editing, small verified steps, evidence before done, lean context, short reports. Use when starting any coding task, when the user types /ds-mode, or when work starts to drift.
Write a numbered plan with an acceptance check for each step, critique it once in a fresh pass, revise it, then execute one step at a time. Use when a task touches more than one file, has unclear scope, or failed before because steps were missed. Invoke as /ds-plan-loop.
Files a short, clear pull request for the current branch. Use when the user asks to open, file or create a PR, or says /ds-pr.
Formats and posts GitHub comments written by an agent on the user's behalf, with clear attribution and no leaked secrets. Use whenever you post a PR comment, review reply or dismissal reason, or on /ds-pr-comments.
Compact engineering principles for design, debugging, verification, and delegation, each with a trigger. Use when choosing structure or types, sizing a diff, debugging, writing tests, or when the user types /ds-principles.
Runs the full ds-ship workflow but stops with the pull request ready for a human to merge, and never merges. Use when the user says get it ready, make it mergeable, stop before merge, or /ds-ready.
Runs an author and a fresh-context reviewer in a loop on the current change until both agree it is correct, for at most four rounds. Use before or after filing a PR, when the user says review loop, get it reviewed, or /ds-review-loop.
Brings an old PR up to date by checking if its idea still fits the current code, then rebasing, reworking or recommending it be closed. Use when asked to rebase, refresh, update or revive a stale PR, or /ds-revive.
Restate the user's goal, make the task deeper rather than bigger, rewrite the prompt, and optionally attack the plan to find weak spots. Use before starting vague or high-stakes work, for 'sharpen this', 'what am I really asking', 'stress test this plan', or /ds-sharpen.
Takes a change all the way from code to a merged pull request without supervision: build it, prove it, review it, file the PR, watch it and merge it. Use when the user says ship it, send it all the way, merge when green, or /ds-ship.
Fan out N parallel workers on slices of a job or on a race, collect every result, and return one evidence-backed report. Use for wide coverage, audits, races between approaches or exploration, or for /ds-swarm, 'swarm this', 'split this across workers'.
Fix a bug by first writing a focused test that fails for the right reason, then making the smallest change that makes it pass. Use when the user asks for TDD or a regression test, when a bug has a cheap local test target, or for /ds-tdd.
Remove AI writing tells from prose and code comments, then rewrite in plain, specific English. Use when writing or editing replies, docs, READMEs, PR descriptions, commit messages, or comments, or when the user types /ds-unslop.
Break out of a repeating failure: name the assumption behind all the failed fixes, test it directly, list three other approaches, pick one, and restart clean if needed. Use when the same error or wrong behavior comes back after two fix attempts, or when you notice you are retrying small variations.…
Run edit, real check, read the actual error, fix the root cause, rerun, until the check passes or the cap is hit. Use when you change code that has tests, a type checker, a build, or a runnable app, or before you claim any fix works. Invoke as /ds-verify-loop.
Watches a pull request through CI and review, fixes real findings and stops when it is ready, merged or blocked. Use when the user asks to watch, monitor or babysit a PR, or /ds-watch-pr.
MCP = 手
AI を外部のサービスやデータにつなぎます(例:Notion を読む、請求書を作る)。プログラムやサーバーが動いている必要があります。
Skill = 手順書
作業のやり方を AI に教えるテキストファイル。どのサービスにもつながりません。
重なる部分もあります。ツールを呼ぶスクリプト付きの Skill や、知識だけを渡す MCP もあります。やりたい作業から選んでください。
Claude CodeCodexOpenCode
提供元の README と repo のフォルダから判断
This repo has 29 skills. Pick one from the list above; its page shows every way to install it, including letting the AI install it and a ZIP
もう入っていますか?先に確認 · ターミナルを開き、このコマンドを入力して Enter
node -vMac:⌘ + Space を押し、Terminal と入力して Enter
Windows:スタートメニューを開き、Windows PowerShell と入力して「開く」
まだ入っていない場合: 公式ダウンロードページで LTS 版と、クリックで進むインストーラーを選びます:Mac は .pkg、Windows は .msi。ダブルクリックして最後まで進めるだけで、curl コマンドは不要です
入れ終わったらターミナルを閉じて開き直し、同じコマンドをもう一度実行します。バージョン番号が出れば完了です: node -v
一度入れれば、Node.js を使うすべてのツールで使えます
コマンドを確認中 · 確認が通るとコピーボタンが出ます
提供元の README から引用。提供元のコードをダウンロードして実行します
質問、コツの共有、使えないことの報告ができます。「役に立った」の多い順です。