Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
作者提供的说明(英文原文)
Dependency policy and MCP server to stop AI coding agents from installing risky npm packages.
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json
{
"mcpServers": {
"pkgdiet": {
"command": "npx",
"args": [
"-y",
"pkgdiet"
]
}
}
}在终端中运行这条命令
claude mcp add pkgdiet -- npx -y pkgdiet粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"pkgdiet": {
"command": "npx",
"args": [
"-y",
"pkgdiet"
]
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"pkgdiet": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"pkgdiet"
]
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。
你的选择通过 Cookie 保存在这台设备上,无需账户。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月7日
注册名称已通过 GitHub 账户 om-tajne 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
它在你的电脑上运行,没有互联网端点
在你的电脑上运行,改用本地设置
它在你的电脑上运行;我们只能读取互联网端点的工具列表
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
GitHub 仓库最后更新于–,最新版本(2.0.1)于7天前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js。
同一分类中有:Spec Driven Development, Remote Desktop Commander, Kanal Varna — drains, leaks & plumbing (Varna, BG)
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Drain unblocking, leak detection & plumbing in Varna, Bulgaria: services, prices, areas, contact
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
Dependency policy for AI-assisted JavaScript and TypeScript development.
*Put your node_modules on a diet.*
The Problem: AI coding agents (Claude Code, Codex, Cursor, Windsurf, Cline) propose nonexistent, deprecated, vulnerable, or typosquatted packages. There is no built-in mechanism to stop them from running the install.
The Solution: PkgDiet intercepts every npm install / pnpm add / yarn add command your AI agent generates — using native agent hook APIs — and blocks unapproved installs *before* they execute.
Run this to see PkgDiet block real risky AI-generated dependencies in your terminal:
npx pkgdiet demo
Example output:
🔒 PkgDiet — Live AI Dependency Enforcement Demo
Simulating what happens when an AI agent tries to install these packages.
🟡 WARNING — moment Context: AI suggested this for date formatting Score: 100/100 Reason: Efficiency Flag: Better alternatives exist for moment. Use this: dayjs, date-fns
🔴 BLOCKED — request Context: AI suggested this for HTTP requests Score: 15/100 Reason: Health score 15 is below minimum allowed (60). Use this: undici, native fetch CVEs: GHSA-p8p7-x288-28g6
🔒 PkgDiet blocked this risky AI-generated dependency before install.
🔴 BLOCKED — node-uuid Context: AI suggested this for UUID generation Score: 15/100 Reason: Health score 15 is below minimum allowed (60). Use this: crypto.randomUUID(), uuid
🔒 PkgDiet blocked this risky AI-generated dependency before install.
🟡 WARNING — lodash Context: AI suggested this for array utilities Score: 80/100 Reason: Efficiency Flag: Better alternatives exist for lodash. Use this: lodash-es, native JS
📊 Demo Summary: 2 blocked, 2 warned These checks run automatically before every npm install in your AI agent. Setup: npx pkgdiet init
For machine-readable CI proof:
npx pkgdiet demo --json
npx pkgdiet init
This single command creates:
| File | Purpose |
|---|---|
.pkgdietrc.json | Repository dependency policy |
.github/workflows/pkgdiet.yml | GitHub Actions CI gate |
.cursor/mcp.json | Cursor MCP server |
.cursor/rules/pkgdiet.mdc | Cursor always-applied enforcement rule |
.cursorrules | Cursor advisory rule (legacy) |
.windsurfrules | Windsurf advisory rule |
cline_mcp_settings.json | Cline MCP server |
.github/mcp.json | GitHub Copilot MCP |
CLAUDE.md | Claude Code advisory rule |
.claude/settings.json | Claude Code PreToolUse hook registration |
.claude/hooks/pkgdiet-install-guard.sh | Claude Code bash install guard (exit code 2) |
.codex/hooks.json | Codex PreToolUse hook registration |
.codex/hooks/pkgdiet-install-guard.mjs | Codex Node.js install guard |
PkgDiet uses each agent's native hook API to block installs — not just advisory rules:
| Agent | Enforcement Mechanism | What it blocks |
|---|---|---|
| Claude Code | PreToolUse bash hook (.claude/hooks/) | npm/pnpm/yarn install before execution |
| Codex (OpenAI) | PreToolUse Node.js hook (.codex/hooks/) | npm/pnpm/yarn install before execution |
| Cursor | MDC alwaysApply rule + MCP check_dependency | Prevents AI from writing install commands |
| Windsurf | .windsurfrules (advisory rules only) | Guides AI to avoid prohibited packages |
| Cline | MCP check_dependency | Blocks before install via tool call |
| GitHub Copilot | MCP check_dependency | Blocks before install via tool call |
| Antigravity | Native MCP server (.gemini/antigravity/mcp/) | Contextual tool evaluations |
| CI (GitHub Actions) | pkgdiet ci gate | Blocks PR merge if any new package fails policy |
How PreToolUse hooks work: When Claude Code or Codex generates a Bash tool call containing
npm install, the registered hook script runs first. If PkgDiet returns a non-zero exit code (exit code 2 for Claude Code), the agent sees the denial reason and aborts the install — the command never runs.
PkgDiet is packaged as a Claude Code plugin featuring bundled MCP tools, slash commands (/pkgdiet:check, /pkgdiet:audit), skill definitions, and PreToolUse execution hooks:
/plugin install github:om-tajne/pkgdiet
*(Or /plugin install pkgdiet@claude-plugins-official once listed in the official directory)*
PkgDiet uses a shared core engine (@pkgdiet/core) so all interfaces apply the same evaluation logic.
.pkgdietrc.json file dictates what is allowed, warned, or blocked for your project.check_dependency instructs other agents before they write the command.npx pkgdiet ci runs in GitHub Actions, diffs package.json against the base commit, and fails the PR if any added package violates policy.| Command | Description |
|---|---|
| `audit` | Audit existing dependencies for policy, health, size, and unused-package signals |
| `check ` | Evaluate npm packages against this repository's dependency policy (ALLOW, WARN, BLOCK) |
| `demo` | Demonstrate local policy evaluation against simulated risky package installs |
| `mcp` | Start the local MCP JSON-RPC server over stdio for MCP-compatible AI coding agents |
| `ci` | Enforce policy for dependency changes introduced by this branch (lockfile-diff-based) |
| `init` | Set up PkgDiet in this project — creates policy, CI workflow, and all AI agent configs |
| `setup` | Interactive wizard for creating a starter .pkgdietrc.json policy |
| `agent-setup` | Configure PkgDiet for AI coding agents and supported install guards |
| `alternatives` | Browse the PkgDiet alternatives dataset |
| `drift` | Scan project for dependency health drift over time |
| `pr` | Generate a reviewer-ready pull request template for adding PkgDiet to your repository |
| `policy-check` | Validate the repository's .pkgdietrc.json policy |
| `cache` | Manage the local PkgDiet registry cache |
.pkgdietrc.json){
"minHealthScore": 70, "warnHealthScore": 80, "blockDeprecated": true, "blockKnownVulnerabilities": true, "blockTyposquats": false, "requirePinnedVersions": true, "maxPackageSizeBytes": 15728640, "failOn": "BLOCK", "securityMode": "fail-open", "blockedPackages": ["moment", "request"], "internalNamePrefixes": [], "environments": { "ci": { "minHealthScore": 80, "failOn": "BLOCK", "securityMode": "fail-closed" }, "dev": { "minHealthScore": 60, "failOn": "BLOCK", "securityMode": "fail-open" }, "staging": { "minHealthScore": 70, "failOn": "BLOCK", "securityMode": "fail-open" } } }
PkgDiet acts as a local Model Context Protocol (MCP) server over stdio.
When your AI coding agent connects to PkgDiet, it gains access to 5 read-only tools:
check_dependency: Evaluates an npm package against your local .pkgdietrc.json policy and returns structured ALLOW, WARN, or BLOCK verdicts with evidence.check_dependencies: Bounded batch evaluation of up to 50 dependencies in a single tool call.suggest_alternative: Queries PkgDiet's curated dataset to find modern, lighter, and maintained alternatives for blocked packages.get_policy: Retrieves the active repository dependency policy rules and thresholds.explain_verdict: Detailed breakdown of package scoring, maintenance signals, and diagnostic codes.Built-in Verification (all free, zero paid API keys):
PKGDIET_NO_NETWORK=1: Runs completely offline using cached metadata; never contacts the external registry.PKGDIET_TELEMETRY_DISABLED=1: Disables local telemetry and metrics collection.PKGDIET_REGISTRY_URL: Overrides the default npm registry with a custom or private enterprise registry endpoint..pkgdiet-cache.json: Local cache file storing validated package metadata and advisory results..pkgdiet-metrics.json: Local metrics file for tracking audit performance and check history.| Package | Path | Purpose |
|---|---|---|
| `@pkgdiet/core` | packages/core | Deterministic policy evaluation engine, OSV matcher, and cache manager |
| `@pkgdiet/mcp` | packages/mcp | Model Context Protocol server exposing check_dependency and alternatives |
| `pkgdiet` | packages/cli | Unified command-line interface, agent setup wizards, and CI gate |
| GitHub Action | action.yml | Reusable GitHub Action for pull request dependency gating |
只显示了开头部分;其余内容请在仓库中阅读