Readyorai

io.github.TedoNeObichaJavaScript由 GitHub 账号 TedoNeObichaJavaScript 发布v0.1.2更新于 Oct 24, 2026

Zero-cost MCP server for local code inspection and best-practice recommendations

已验证STDIO仅桌面开发者工具安全与监控

概览

AI 生成的概览

本地零成本 MCP 服务器，用于检查源代码的质量、安全与最佳实践问题并给出修复建议。

可以做什么
  • 检查单个文件或整个目录的代码质量问题
  • 获取代码行数、复杂度、函数数量等量化指标
  • 比较两个文件版本，查看已修复、新引入或未变化的发现
  • 针对每条发现获取带上下文的可执行修复建议
  • 运行结构化的代码审查与项目健康检查提示
适用场景
适合希望让助手在本地审查代码的整洁度、复杂度、命名、重复与安全模式，而不把代码发送到外部服务的场景，也适合对照基线跟踪逐步清理问题。
运行要求
以本地 stdio 进程运行，通常通过 npx 从 readyorai npm 包启动，因此需要 Node.js 与 npm。未声明任何 API 密钥或环境变量，分析完全在本地进行。
安装前请注意
服务器会读取本地源代码文件，请只指向你愿意让助手访问的目录。它只报告问题，不会修改代码。项目配置文件与行内抑制注释会影响显示哪些发现。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Readyorai，将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only，通过 STDIO。 STDIO 服务会启动本地进程，因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

ReadyOrAI

Zero-cost MCP server for local code inspection. Analyzes your code for clean code practices, best practices, and gives actionable recommendations — without modifying your code or making API calls.

Features

  • 9 analyzers: complexity, naming, structure, patterns, imports, documentation, security, duplication, AI-generated code detection
  • Multi-language: JS/TS, Python, Go, Rust (deep analysis) + Java, C#, Ruby, PHP, and more (regex-based)
  • Zero-cost: All analysis runs locally — no API keys, no external calls
  • MCP server: Works with Claude Desktop, Claude Code, and VS Code via stdio transport
  • CLI tool: ready @filename and AI terminal commands for standalone use
  • SARIF output: Export results for GitHub Code Scanning integration
  • Watch mode: Continuous re-analysis on file changes
  • Baseline/diff mode: Track incremental adoption — only see new issues
  • Config file: Project-level .readyorai.json for shared settings
  • Inline suppressions: Silence specific findings with // readyorai-ignore comments

Installation

As an MCP Server

Claude Desktop (claude_desktop_config.json):

json
{  "mcpServers": {    "readyorai": {      "command": "npx",      "args": ["-y", "readyorai"]    }  }}

Claude Code:

bash
claude mcp add readyorai -- npx -y readyorai

VS Code (.vscode/mcp.json):

json
{  "servers": {    "readyorai": {      "command": "npx",      "args": ["-y", "readyorai"]    }  }}

As a CLI Tool

bash
npm install -g readyorai

CLI Usage

bash
ready @src/index.ts                          # Inspect a single fileready @src/                                  # Inspect a directoryready @src/index.ts --checks security,naming # Run specific checksready @src/index.ts --json                   # JSON outputready @src/index.ts --sarif                  # SARIF output (GitHub Code Scanning)ready @src/index.ts --severity warning       # Only warnings and errorsready @src/index.ts --watch                  # Re-analyze on file changesready @src/ --baseline baseline.json         # Compare against saved baselineready @src/ --save-baseline baseline.json    # Save current results as baselineAI                                           # Inspect current directoryAI @src/utils.ts                             # Alias with file target

Exit Codes

CodeMeaning
0No issues found
1Warnings found
2Errors found
3Runtime error

Configuration

Create a .readyorai.json in your project root:

json
{  "checks": ["complexity", "naming", "security"],  "severity": "warning",  "exclude": ["dist/**", "node_modules/**"]}

Inline Suppressions

Silence specific findings on a line:

js
eval(code); // readyorai-ignore

MCP Tools

ToolDescription
inspect_fileAnalyze a single file for code quality issues
inspect_directoryBatch analysis of all source files in a directory
get_metricsGet quantitative metrics (LOC, complexity, function count)
compare_filesCompare two file versions and show fixed, introduced, or unchanged findings
suggest_fixesGet actionable fix suggestions with context for each finding

MCP Prompts

PromptDescription
review-codeStructured code review with optional focus area
health-checkProject-level health assessment

Checks

CheckWhat it detects
complexityCyclomatic/cognitive complexity, deep nesting
namingConvention violations, single-letter vars, boolean prefixes
structureLong files/functions, too many parameters, long lines
patternsconsole.log, empty catch, magic numbers, nested ternaries, TODOs
importsUnused imports, wildcard imports, scattered imports
documentationMissing JSDoc/docstrings, low comment ratio
securityHardcoded secrets, eval(), SQL injection, XSS patterns
duplicationDuplicate code blocks, repeated magic strings
ai-detectionDetects patterns common in AI-generated code

Language Support

TierLanguagesAnalysis
1JavaScript, TypeScript, Python, Go, RustFull AST-powered analysis
2Java, C#, Ruby, PHP, Swift, Kotlin, C, C++Regex + heuristic analysis
3Any text fileLine-based checks (length, TODOs, secrets, duplication)

Development

bash
git clone https://github.com/TedoNeObichaJavaScript/ReadyOrAI.gitcd ReadyOrAInpm installnpm run buildnpm test

Support

If you find ReadyOrAI useful, consider supporting the project:

  • Borko — 5 euro

License

MIT

来源：README.md，提交 e7d0593

工具

0
SourceWeft 尚未在沙箱中启动过该软件包，因此其工具未知。

版本历史

1
  1. v0.1.2最新Sep 16, 2026

评分

已满足适用检查项的 50%

该评分衡量此收录条目的完整程度，依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价，也不表示服务器安全或当前可用：可连通性是带日期的检查结果，评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分

    请保持端点能够响应 MCP 握手；SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头，或保持端点能够响应探测，以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list，或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分