Qartez Mcp

Name: Qartez Mcp
Author: kuberstar

@kuberstar

代码仓库主页

关于 Qartez Mcp

Semantic code intelligence MCP server for Claude Code - project maps, symbol search, impact analysis, and more

#code-analysis #developer-tools #tree-sitter #rust

基本信息

分类

其他

许可证

NOASSERTION

运行时

rust

传输方式

stdio

发布者

kuberstar

提交者

KUBERSTAR

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "qartez": {
      "command": "qartez-mcp",
      "args": []
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Qartez Mcp?

Qartez Mcp is a code-intelligence MCP server that builds a pre-computed knowledge graph of a repository—symbols, imports, call edges, PageRank, git co-change, cyclomatic complexity—and serves it to AI assistants through the Model Context Protocol. It is designed to be consumed by language models rather than humans, cutting AI token usage by approximately 94% compared to traditional tools like grep, find, cat, and ls.

How to use Qartez Mcp?

Install via the one-liner: curl -sSfL https://qartez.dev/install | sh (macOS/Linux/WSL2) or the PowerShell equivalent for Windows. The installer places three binaries (qartez, qartez-guard, qartez-setup) and automatically configures every MCP-capable IDE it detects. Once installed, open any project; Qartez indexes it automatically on session start. Manual re-indexing is available with qartez --root /path/to/project --reindex. The file watcher keeps the index fresh as you edit.

Key features of Qartez Mcp

Pre-computed knowledge graph of the entire codebase
43 MCP tools for querying symbols, impact, and structure
Supports 37 programming languages
Reduces AI token consumption by ~94% vs. grep-based approaches
Includes modification guard hooks for Claude Code
Works with 19 editors and agents out of the box
Platform support: macOS 13+, Ubuntu 22.04+, Windows, WSL2 (x86_64 and arm64)

Use cases of Qartez Mcp

Quickly find where a symbol is defined without scanning files
Assess the blast radius before editing an important file
Identify high-risk code via hotspot analysis (complexity × PageRank × churn)
Detect dead exports and structural code clones across the repository
Enforce architecture boundaries by declaring import rules

FAQ from Qartez Mcp

What does Qartez Mcp do differently from grep or find?

It builds a pre-computed knowledge graph once and serves it through MCP, so the AI assistant queries the index instead of scanning files byte-by-byte. This reduces token usage by ~94% and avoids re-reading the same directories on every question.

What editors and agents does it support?

Claude Code, Claude Desktop, Gemini, Cursor, Windsurf, Kiro, Zed, Continue.dev, Copilot CLI, Amazon Q, Amp, Cline, Roo Code, Goose, Warp, Augment, OpenCode, Codex CLI, and Antigravity.

What languages does Qartez Mcp support?

37 languages are supported (the README does not list them individually, but the badge confirms the number).

How do I install Qartez Mcp?

Use the one-liner curl -sSfL https://qartez.dev/install | sh (macOS/Linux/WSL2) or the PowerShell script for Windows. Alternatively, clone the repository and run make deploy. The installer downloads pre-built binaries, verifies SHA-256, installs three binaries, and configures detected IDEs.