fuck-u-code is a feature-rich code quality analysis tool with multi-language and AI review support. It is well-suited for developers who enjoy a humorous tone and want quick insights into code quality. However, privacy risks from AI review and its informal nature may not fit all teams.
Strongest · Setup experience 12/20 Weakest · Reliability 5/20
Static review · not runListed 2026-08-07
Read the FMRS scoring method →Runtime：Node.js
OPENAI_API_KEY
API key for OpenAI or compatible providers, needed only for ai-review with the openai provider; obtain it from the OpenAI platform.
ANTHROPIC_API_KEY
Anthropic API key, needed only for ai-review with the anthropic provider; obtain it from the Anthropic console.
DEEPSEEK_API_KEY
DeepSeek API key, needed only for ai-review with the deepseek provider; obtain it from the DeepSeek platform.
GEMINI_API_KEY
Gemini API key, needed only for ai-review with the gemini provider; obtain it from Google AI Studio.
OPENAI_BASE_URL
Optional custom OpenAI-compatible API endpoint.
OPENAI_MODEL
Optional default model name.
OLLAMA_HOST
Optional local Ollama host address, set when using the ollama provider for local review.
Install globally: npm install -g eff-u-code. Then configure MCP automatically: fuck-u-code mcp-install (interactive) or specify client: fuck-u-code mcp-install claude / cursor. For manual configuration, add an mcpServers entry to your client config file. For example, in Claude Code's ~/.claude.json or project .mcp.json, add {"mcpServers":{"fuck-u-code":{"command":"fuck-u-code-mcp"}}}. Alternatively, you can use npx: set command to "npx" and args to ["-y", "eff-u-code-mcp"].
{
"mcpServers": {
"fuck-u-code": {
"command": "fuck-u-code-mcp"
}
}
}
Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.
{
"servers": {
"fuck-u-code": {
"command": "fuck-u-code-mcp"
}
}
}
Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).
claude mcp add fuck-u-code -- fuck-u-code-mcp
Run it in a terminal; replace any <…> placeholders with your own values first.
The client's tool list should show analyze and ai-review; issuing a request like 'analyze the code quality of the current directory' and receiving a scored report confirms the connection works.
Once connected, you can ask your AI assistant things like:
No matching tools
Listed from the project's documentation, not tested by this site.
fuck-u-code is a code quality analysis tool designed to expose shitty code quality with sharp but humorous feedback. It uses AST parsing via tree-sitter for accurate syntax analysis, supports 14 programming languages, and performs seven quality checks: complexity, size, comments, error handling, naming, duplication, and structure. Provides an overall score from 0 to 100 (higher is better) and a per-file Shit-Gas Index (higher is worse). Supports AI code review with OpenAI-compatible, Anthropic, DeepSeek, Gemini, and Ollama providers, multiple output formats (colored terminal, Markdown, JSON, HTML), and i18n (English, Chinese, Russian). As an MCP server, it can be invoked by AI tools like Claude Code, Cursor, Windsurf, etc. Code analysis runs fully offline, so your code never leaves your machine.
Upstash's official server providing up-to-date third-party library docs for AI coding assistants
AI-powered NPM package analysis MCP server
An independent control layer that gives AI coding agents budgets, brakes, and receipts
GitHub's official MCP server for managing repos, issues, PRs, and workflows via natural language
Source revision 27ae48f50816 Data synced 2026-10-11 Read the FMRS scoring method