MCP Catalogs
Home

ckb

by nyxCore-Systems·95·Score 50

CKB is a comprehensive code intelligence platform providing MCP server with tools for code analysis, impact assessment, and architecture mapping.

developer-toolsai-llmknowledge-graph
11
Forks
10
Open issues
1 mo ago
Last commit
2d ago
Indexed

Overview

CKB transforms codebases into queryable knowledge bases, offering semantic search, call graphs, impact analysis, and architecture mapping. It provides compound operations that reduce AI tool calls by 60-70% and integrates with AI assistants like Claude Code, Cursor, and Windsurf via MCP. The system supports multiple interfaces including CLI, HTTP API, and MCP, with language support for Go, TypeScript, JavaScript, Python, and more.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:AI-assisted code review with precise impact analysis and risk scoring
you:Architecture understanding and documentation through semantic queries
you:Dead code detection and tech debt identification in large codebases
you:What programming languages does CKB support?
you:How does CKB integrate with AI assistants?

When to choose this

Choose CKB when you need deep code intelligence for AI-assisted development, especially for Go codebases or when working with TypeScript/JavaScript/Python projects that require impact analysis and architecture mapping.

When NOT to choose this

Avoid CKB if you're working with unsupported languages (C#, PHP) or need non-Go incremental indexing, as the tool's effectiveness varies significantly by language tier and reindexing capabilities.

Tools this server exposes

12 tools extracted from the README
  • explore

    Area exploration: finds symbols, dependencies, and hotspots in a code area

  • understand

    Symbol deep-dive: finds references, callers, and explanations for a specific symbol

  • prepareChange

    Pre-change analysis: determines impact, affected tests, and risk of changing code

  • batchGet

    Fetch up to 50 symbols at once for efficient bulk operations

  • batchSearch

    Run up to 10 searches at once to reduce API overhead

  • findReferences

    Find all references to a specific symbol in the codebase

  • searchSymbols

    Search for symbols by name with semantic understanding

  • hotspots

    Identify risky areas in the codebase based on complexity, coupling, and change frequency

  • architecture

    Generate architecture overview and module dependencies

  • ownership

    Find code ownership and reviewer suggestions based on CODEOWNERS and git history

  • findDeadCode

    Detect unused code with confidence scores based on call graphs and references

  • compareAPI

    Analyze API changes between versions to identify breaking changes and affected clients

Comparable tools

code-mcpcodeintel-mcpsourcegraph-mcp

Installation

Installation

Option 1: npm (Recommended)

npm install -g @tastehub/ckb

Option 2: Homebrew (macOS/Linux)

brew tap SimplyLiz/ckb
brew install ckb

Option 3: Build from Source

git clone https://github.com/SimplyLiz/CodeMCP.git
cd CodeMCP
go build -o ckb ./cmd/ckb

Setup for Claude Desktop

Add this to your Claude Desktop config.json:

{
  "mcpServers": {
    "ckb": {
      "command": "npx",
      "args": ["@tastehub/ckb", "mcp"]
    }
  }
}

FAQ

What programming languages does CKB support?
CKB supports Go (Tier 1), TypeScript/JavaScript/Python (Tier 2), Rust/Java/Kotlin/C++/Ruby/Dart (Tier 3), and experimental support for C#/PHP (Tier 4).
How does CKB integrate with AI assistants?
CKB provides an MCP server that exposes tools for code analysis, impact assessment, and architecture mapping to AI assistants like Claude Code, Cursor, and Windsurf, reducing token usage by 60-70% through compound operations.

Compare ckb with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.