devscope
Enables AI agents to detect development environments, install missing tools, scan local code projects, and generate visual reports.
README
devscope
Development environment at a glance — detect, install, and manage your dev tools. Scan & visualize all local code projects. MCP-native.
Why devscope?
Setting up a new machine or onboarding onto a team? devscope answers three questions instantly:
- What's installed? — Detect runtimes, package managers, and version managers in seconds.
- What's missing? — One command to install everything you need.
- What's on this machine? — Scan all your code projects, see languages, frameworks, Git status, and dependencies at a glance.
It also works as an MCP Server, so AI agents (Cursor, Claude Desktop, etc.) can query your dev environment directly.
Features
- Environment Detection — Node, Go, Python, Rust, Java, Ruby, PHP, .NET, Swift, Dart + package managers + version managers
- One-Click Setup — Interactively install missing tools via Homebrew / apt / winget / official installers
- Project Scanner — Discover all code projects across your machine, extract metadata, frameworks, Git info
- Visual Dashboard — Self-contained HTML report with Environment / Projects / Insights views
- MCP Server — AI agents can detect environments, install tools, scan projects, and generate reports
- Beautiful CLI — Colorful terminal output with progress indicators
Quick Start
# No install needed — run directly
npx devscope env # Show what's installed
npx devscope setup # Install missing tools
npx devscope scan # Scan your projects
npx devscope list # List all projects
npx devscope report # Generate visual report
Or install globally:
npm install -g devscope
CLI Commands
| Command | Description |
|---|---|
devscope env |
Detect installed runtimes, package managers, version managers |
devscope setup |
Interactively install missing tools |
devscope setup --all |
Install all missing tools without prompting |
devscope setup --only node,go |
Install specific tools |
devscope scan [dirs...] |
Scan directories for code projects |
devscope list |
List scanned projects |
devscope list --lang node |
Filter by language |
devscope list --active |
Show only active projects |
devscope show <name> |
Show detailed info for a project |
devscope report |
Generate HTML report and open in browser |
devscope mcp |
Start as MCP Server |
MCP Integration
devscope works as an MCP Server for AI-powered development tools.
Cursor
Settings → MCP → Add:
{
"mcpServers": {
"devscope": {
"command": "npx",
"args": ["-y", "devscope", "mcp"]
}
}
}
Claude Desktop
Edit ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"devscope": {
"command": "npx",
"args": ["-y", "devscope", "mcp"]
}
}
}
MCP Tools
| Tool | Description |
|---|---|
detect_environment |
Detect all installed runtimes and tools |
install_tool |
Install a specific runtime or package manager |
scan_directories |
Scan directories and discover projects |
generate_report |
Generate visual HTML report |
Example Prompts
"What development tools do I have installed?"
"Install Go and Rust on my machine"
"Scan ~/workspace and show me all my projects"
"Generate a report of all my code projects"
Detected Technologies
Runtimes
Node.js · Go · Python · Rust · Java · Ruby · PHP · .NET · Swift · Dart
Package Managers
npm · yarn · pnpm · bun · cargo · pip · uv · composer · gem · Homebrew
Version Managers
nvm · fnm · volta · n · pyenv · rustup · mise · asdf · sdkman · rbenv
Project Languages
Node.js (package.json) · Python (requirements.txt, pyproject.toml) · Go (go.mod) · Java (pom.xml, build.gradle) · Rust (Cargo.toml)
Frameworks
React · Vue · Angular · Svelte · Next.js · Nuxt · Electron · Tauri · Express · NestJS · Fastify · Koa
Output
| File | Path | Description |
|---|---|---|
| JSONL | ~/.devscope/results.jsonl |
Raw scan data |
| HTML | ~/.devscope/report.html |
Visual report |
Contributing
See CONTRIBUTING.md for guidelines.
License
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。