GitHub
GitHub’s official server lets AI read and manage code, issues and pull requests in your account. Use GitHub’s hosted version or run it locally with Docker.
Description from the maker (original English)
Auto-detect stubbable packages for Python exe builds and generate minimal stub code
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Open Claude Desktop → Settings → Developer → Edit Config, and paste this into claude_desktop_config.json
{
"mcpServers": {
"pystub": {
"command": "uvx",
"args": [
"mcp-pystub"
]
}
}
}Run this in your terminal
claude mcp add pystub -- uvx mcp-pystubPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"pystub": {
"command": "uvx",
"args": [
"mcp-pystub"
]
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"pystub": {
"type": "stdio",
"command": "uvx",
"args": [
"mcp-pystub"
]
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
This one runs on your computer, and ChatGPT can only connect to servers on the internet. Pick another app.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Runs on your computer; the tool list appears in your AI app after you install it.
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 7, 2026
The registry name was verified with the GitHub account KoyoYeager, which owns the repo
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
This one runs on your computer; there is no internet endpoint
Runs on your computer and uses local settings instead
Runs on your computer; we can only read tool lists from internet endpoints
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
The registry does not list a required key, though some servers ask you to sign in on first use.
It runs on your computer. Your AI app starts it when needed, and data does not pass through the maker’s servers (unless it calls an online service itself).
The GitHub repo last changed –, and the latest version (0.1.1) reached the registry 7 months ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Python (uv) on your computer.
In the same category: GitHub, Desktop Commander, Cloudflare MCP
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
GitHub’s official server lets AI read and manage code, issues and pull requests in your account. Use GitHub’s hosted version or run it locally with Docker.
MCP server for terminal commands, file operations, and process management
Cloudflare MCP servers
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
Python exe ビルド(PyInstaller / Nuitka / cx_Freeze)時にスタブ置換可能なパッケージを自動検出し、最小スタブコードを生成する MCP サーバー。
An MCP server that auto-detects stubbable packages for Python exe builds (PyInstaller / Nuitka / cx_Freeze) and generates minimal stub code to reduce executable size.
Python アプリを exe 化すると、依存ライブラリがモジュールレベルで import する重量パッケージが丸ごと同梱され、exe サイズが膨張する。実際のコードパスで使わないパッケージは最小ダミー(スタブ)で置換すれば大幅なサイズ削減が可能。
When building Python apps into executables, heavy packages imported at module level by dependencies get bundled entirely, bloating the exe size. Replacing unused packages with minimal stubs can significantly reduce size.
asammdf プロジェクト — PyInstaller exe ビルド(E2E 動作確認済み):
| exe サイズ | 動作 | PySide6 | |
|---|---|---|---|
| スタブなし | 431 MB | OK | ロード済み |
| asammdf.gui スタブ適用 | 259 MB | OK | 排除 |
| 削減 | -40%(172 MB) | 影響なし |
asammdf 変換ツール(analyze 結果):
stubbable: pandas (59.7 MB), canmatrix (4.0 MB) — 合計 63.7 MB
submodule hints: asammdf.gui → PySide6 (523 MB) 排除可能 — 50 hints 検出
手動でスタブ化していた 3 パッケージ(pandas, canmatrix, asammdf.gui)を全て自動検出
| ツール / Tool | 説明 / Description |
|---|---|
analyze | import グラフを解析しスタブ候補を自動検出。C拡張パッケージの間接排除ヒントも出力 / Auto-detect stubbable packages + submodule stub hints for C-extension elimination |
graph | import グラフをノード・エッジで可視化 / Visualize import graph as nodes and edges |
check | 特定パッケージの使用状況を詳細分析 / Deep analysis of a specific package's usage |
generate | パッケージスタブの最小コードを生成 + ビルド手順出力 / Generate minimal stub code + build instructions |
generate_submodule | C拡張パッケージを間接排除するサブモジュールスタブを生成 / Generate submodule stubs to indirectly eliminate C-extension packages |
| 判定 / Verdict | 意味 / Meaning |
|---|---|
| stubbable | スタブ化可能。プロジェクトの実行パスで未使用 / Safe to stub. Not used in project's runtime path |
| nofollow | try/except 保護あり。--nofollow-import-to で除外推奨 / Protected import. Use --nofollow-import-to |
| required | スタブ化不可。実際に使用されている / Cannot stub. Actually used at runtime |
mdf.get() は使うが mdf.to_dataframe() は使わない → pandas は stubbableisinstance(x, pd.DataFrame) は stub-safe、pd.DataFrame(data) は実使用class User(BaseModel) → pydantic は requirednofollow と判定generate_submodule で間接排除が可能(v0.2 で対応)importlib.import_module(変数))は静的解析で追跡不可(warnings で通知)__init__ で直接呼ばず後のメソッドで使うパッケージは検出精度が下がる(安全側で required に判定)python-dateutil → dateutil 等のマッピングは未対応if sys.version < (3,11) 内の import は静的解析で判定不可「stubbable と判定したが実は必要だった」は絶対に起こさない設計。判定に迷う場合は required(安全側)に倒す。逆方向の誤判定(本当は stubbable なのに required)は許容する。
ast.parse() で各ファイルの import 文を解析(module-level / function-level / try-except 保護を区別)pip install mcp-pystub
mcp>=1.0.0 - Model Context Protocol SDKmcp-pystub
{
"mcpServers": { "pystub": { "command": "mcp-pystub" } } }
入力 / Input:
entry_point: "C:/project/converter.py" python_path: "C:/project/.venv/Lib/site-packages"
出力 / Output: { "stubbable": [ {"package_name": "pandas", "estimated_size_mb": 59.7, "reason": "依存ライブラリ経由でのみ import..."} ], "required": [ {"package_name": "numpy", "reason": "プロジェクトコードが直接 import し使用"}, {"package_name": "PySide6", "estimated_size_mb": 523.2, "reason": "C 拡張...ただし asammdf.gui をスタブ化することで間接排除が可能", "submodule_stubs": [{"submodule": "asammdf.gui", "target_package": "PySide6"}]} ], "nofollow": [ {"package_name": "mpmath"} ], "submodule_stub_hints": [ {"submodule": "asammdf.gui", "parent_package": "asammdf", "target_package": "PySide6", "imported_symbols": ["plot"]} ], "analysis_time_ms": 6478 }
入力 / Input:
entry_point: "C:/project/converter.py" package_name: "pandas"
出力 / Output: { "files": { "pandas/init.py": "...", "pandas/core/api.py": "class DataFrame: pass\nclass Series: pass\n..." }, "original_file_count": 2980, "stub_file_count": 266, "stub_total_bytes": 209530, "build_instructions": { "install_commands": ["pip uninstall -y pandas", "# cp stubs to site-packages"], "uninstall_commands": ["pip install pandas"], "hook_disable": ["# hook-pandas*.py → .disabled"] } }
入力 / Input:
entry_point: "C:/project/converter.py" parent_package: "asammdf" submodule: "asammdf.gui"
出力 / Output: { "parent_package": "asammdf", "submodule": "asammdf.gui", "files": { "asammdf/gui/init.py": "\"\"\"Auto-generated stub...\"\"\"\ndef plot(*args, **kwargs): ..." }, "eliminated_packages": ["PySide6", "scipy", "lxml"], "original_size_bytes": 5907331, "stub_size_bytes": 144, "build_instructions": { "backup_commands": ["cp -r .../asammdf/gui .../asammdf/gui.bak"], "install_commands": ["rm -rf .../asammdf/gui", "cp -r _stubs/asammdf/gui/ .../asammdf/gui/"], "uninstall_commands": ["mv .../asammdf/gui.bak .../asammdf/gui"], "verify_commands": ["python -c \"import asammdf; print('OK')\""] } }
| プロジェクト規模 | 解析時間 | ノード数 |
|---|---|---|
| 軽量 (click) | 335ms | 73 |
| 中規模 (flask) | 1,743ms | 230 |
| 重量級 (pandas) | 4,820ms | 491 |
| 超重量級 (sympy) | 6,919ms | 681 |
python -m pytest tests/ -v
| テスト種別 | 件数 | 結果 |
|---|---|---|
| ユニットテスト(9モジュール) | 91 | 全通過 |
| PyPI ライブラリ大規模テスト(requests, flask, pandas 等) | 84 | クラッシュ 0 |
| PyInstaller exe ビルド + 動作テスト | 22 | 全 PASS |
| asammdf E2E(MDF作成→読取→リサンプリング) | 1 | exe 40% 削減 + 正常動作 |
| 復元テスト(バックアップ → 復元 → 検証) | 4 | 全成功 |
MIT