Codemcp
CKB is a tool that transforms a codebase into a queryable knowledge base. It helps developers understand code, analyze impacts, find owners, and detect dead code via CLI, API, or AI assistants, improving development efficiency and security.
2.5 points
0

What is CKB?

CKB (Code Knowledge Backend) is a code intelligence platform that analyzes your codebase to build a comprehensive knowledge graph, enabling you to ask code - related questions as if you were consulting a senior engineer. It understands the code's structure, dependencies, ownership, and risks, helping you make code changes safely and release with confidence.

How to use CKB?

CKB offers three main usage methods: 1) Integrate with AI assistants (such as Claude Code, Cursor) via the MCP protocol and ask questions directly in the IDE; 2) Use the command - line tool for quick queries and analysis; 3) Build custom tools or integrate into the CI/CD process via the HTTP API. The core steps include initializing the project, generating the code index, and then you can start querying.

Use cases

CKB is particularly suitable for the following scenarios: providing context when developers use AI assistants for coding, helping teams navigate and refactor in large codebases, understanding the full impact of changes during code reviews, allowing technical leaders to track architecture health, and any situation where code changes need to be made safely.

Main Features

Compound Operations
Provides single - call tools such as `explore`, `understand`, `prepareChange`, which combine multiple queries and can reduce AI tool calls by 60 - 70%, significantly improving efficiency.
Code Navigation and Discovery
Supports semantic search, call graphs, usage tracking, and finding entry points, allowing you to quickly understand the code structure and relationships.
Impact Analysis and Security
Calculates the 'blast radius' of changes, performs risk scoring, detects affected tests, and warns of destructive changes, helping you assess risks before making changes.
Architecture Understanding
Provides module overviews, architecture decision records (ADR), dependency graphs, and explains the origin and design intent of the code.
Ownership and Review
Determines code owners by combining the CODEOWNERS file and Git history (blame), provides reviewer suggestions, and detects ownership drift.
Code Quality and Risk
Detects dead code, analyzes module coupling, identifies code hotspots and high - complexity areas, helping to discover technical debt.
Security Scanning
Automatically scans the code for exposed keys and credentials, supports 26 common patterns, and provides an allow - list function.
Documentation Intelligence
Links documentation to code symbols, detects outdated documentation, and provides documentation coverage metrics.
Multi - Repository and Federated Query
Supports querying across multiple code repositories, analyzes API contracts, and can provide indexing services remotely.
Runtime Intelligence
Integrates OpenTelemetry and combines observed actual usage to help identify dead code in the production environment.
Streaming Response
Provides server - sent events (SSE) streams for operations such as `findReferences` and `searchSymbols`, supporting real - time progress feedback.
Automation and CI/CD
Supports daemon mode, monitoring mode, Webhook triggering, and incremental indexing, facilitating integration into automated processes.
Advantages
Supercharge AI assistants: Provide in - depth code context for AI coding tools such as Claude Code and Cursor, making their answers more accurate.
Safe refactoring: Precisely understand the scope and risks of changes before modifying code to avoid accidental breakage.
Efficient navigation: Quickly find relevant code, understand dependencies, and ownership in large and complex codebases.
Reduce cognitive load: Transform code knowledge into queryable data, eliminating the need for developers to remember all details.
Flexible integration methods: Support CLI, HTTP API, and MCP protocol, adaptable to various workflows and tools.
Intelligent preset optimization: Through toolset presets, the context token usage can be reduced by up to 83%.
Limitations
Mainly static analysis: Based primarily on the static structure of the code, with limited ability to analyze dynamic dispatch and runtime behavior.
Dependent on index quality: The completeness and accuracy of the functions depend on the maturity of the SCIP indexer, and the support for different languages varies.
Cannot handle unindexed generated code: If the generated code is not included in the index, CKB cannot analyze it.
Not a code generation or formatting tool: Focuses on understanding and analysis, does not provide code generation, linting, or formatting functions.
Initial indexing takes time: For large projects, generating a complete initial index may take some time.
Cross - repository calls require additional configuration: By default, cross - repository call analysis requires setting up a federated query.

How to Use

Install CKB
You can install CKB via npm, Homebrew, or by building from source code. npm is the simplest way.
Initialize the Project
Navigate to the root directory of your project and run the initialization command. This will create the necessary configuration files in the project.
Generate the Code Index
Generate a SCIP index for your codebase. This is a key step for accurate code navigation and impact analysis. CKB will automatically detect the language and run the corresponding indexer.
Connect to the AI Assistant
Run the setup command, and CKB will automatically configure the connection with AI tools such as Claude Code for you.
Start Querying
Now you can query directly via the CLI or ask questions about the code directly in the AI assistant session.

Usage Examples

Safely Refactor a Function
You want to rename or modify a core function but are worried about breaking other code that calls it.
Find the Entry Point for a New Feature
You need to add a new feature but are not sure where to start modifying the codebase.
Determine Code Reviewers
You have submitted a PR but are not sure who to invite for review because it involves multiple modules.
Check for Dead Code
You suspect that some old utility functions or classes are no longer in use and want to clean them up.
Understand the System Architecture
You are a new developer joining the project and need to quickly understand how the system is organized.

Frequently Asked Questions

Which programming languages does CKB support?
How often does the index need to be updated?
How does CKB work with my AI assistant (such as Claude Code)?
Will CKB read or send my code data?
What is a 'Preset'? Which one should I use?
Is CKB free?

Related Resources

Official Documentation Wiki
Contains complete documentation such as quick start, detailed feature descriptions, language support, and API references.
Prompt Handbook
Provides real - world question examples and tips to help you better use CKB for asking questions.
Integration Guide
Details how to use CKB's CLI and API in Node.js, Python, Go, and Shell scripts.
GitHub Repository
CKB's source code and issue tracking.
npm Package Page
View the latest version and npm installation instructions.
Official Website
CKB's official website, containing information such as overviews and pricing.

Installation

Copy the following command to your Client for configuration
{
  "mcpServers": {
    "ckb": {
      "command": "npx",
      "args": ["@tastehub/ckb", "mcp"]
    }
  }
}

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

{
  "mcpServers": {
    "ckb": {
      "command": "npx",
      "args": ["-y", "@tastehub/ckb", "mcp"],
      "env": {
        "CKB_REPO": "/path/to/your/repo"
      }
    }
  }
}

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

{
  "mcpServers": {
    "ckb": {
      "command": "npx",
      "args": ["@tastehub/ckb", "mcp", "--preset=review"]
    }
  }
}
Note: Your key is sensitive information, do not share it with anyone.

Alternatives

A
Airweave
Airweave is an open - source context retrieval layer for AI agents and RAG systems. It connects and synchronizes data from various applications, tools, and databases, and provides relevant, real - time, multi - source contextual information to AI agents through a unified search interface.
Python
7.0K
5 points
V
Vestige
Vestige is an AI memory engine based on cognitive science. By implementing 29 neuroscience modules such as prediction error gating, FSRS - 6 spaced repetition, and memory dreaming, it provides long - term memory capabilities for AI. It includes a 3D visualization dashboard and 21 MCP tools, runs completely locally, and does not require the cloud.
Rust
4.5K
4.5 points
B
Better Icons
An MCP server and CLI tool that provides search and retrieval of over 200,000 icons, supports more than 150 icon libraries, and helps AI assistants and developers quickly obtain and use icons.
TypeScript
6.7K
4.5 points
A
Assistant Ui
assistant - ui is an open - source TypeScript/React library for quickly building production - grade AI chat interfaces, providing composable UI components, streaming responses, accessibility, etc., and supporting multiple AI backends and models.
TypeScript
7.3K
5 points
A
Apify MCP Server
The Apify MCP Server is a tool based on the Model Context Protocol (MCP) that allows AI assistants to extract data from websites such as social media, search engines, and e-commerce through thousands of ready-to-use crawlers, scrapers, and automation tools (Apify Actors). It supports OAuth and Skyfire proxy payment and can be integrated into MCP clients such as Claude and VS Code through HTTPS endpoints or local stdio.
TypeScript
7.5K
5 points
R
Rsdoctor
Rsdoctor is a build analysis tool specifically designed for the Rspack ecosystem, fully compatible with webpack. It provides visual build analysis, multi - dimensional performance diagnosis, and intelligent optimization suggestions to help developers improve build efficiency and engineering quality.
TypeScript
10.4K
5 points
N
Next Devtools MCP
The Next.js development tools MCP server provides Next.js development tools and utilities for AI programming assistants such as Claude and Cursor, including runtime diagnostics, development automation, and document access functions.
TypeScript
9.7K
5 points
T
Testkube
Testkube is a test orchestration and execution framework for cloud-native applications, providing a unified platform to define, run, and analyze tests. It supports existing testing tools and Kubernetes infrastructure.
Go
6.5K
5 points
G
Gitlab MCP Server
Certified
The GitLab MCP server is a project based on the Model Context Protocol that provides a comprehensive toolset for interacting with GitLab accounts, including code review, merge request management, CI/CD configuration, and other functions.
TypeScript
24.4K
4.3 points
N
Notion Api MCP
Certified
A Python-based MCP Server that provides advanced to-do list management and content organization functions through the Notion API, enabling seamless integration between AI models and Notion.
Python
20.4K
4.5 points
D
Duckduckgo MCP Server
Certified
The DuckDuckGo Search MCP Server provides web search and content scraping services for LLMs such as Claude.
Python
71.7K
4.3 points
M
Markdownify MCP
Markdownify is a multi-functional file conversion service that supports converting multiple formats such as PDFs, images, audio, and web page content into Markdown format.
TypeScript
35.3K
5 points
U
Unity
Certified
UnityMCP is a Unity editor plugin that implements the Model Context Protocol (MCP), providing seamless integration between Unity and AI assistants, including real - time state monitoring, remote command execution, and log functions.
C#
31.1K
5 points
F
Figma Context MCP
Framelink Figma MCP Server is a server that provides access to Figma design data for AI programming tools (such as Cursor). By simplifying the Figma API response, it helps AI more accurately achieve one - click conversion from design to code.
TypeScript
64.4K
4.5 points
G
Gmail MCP Server
A Gmail automatic authentication MCP server designed for Claude Desktop, supporting Gmail management through natural language interaction, including complete functions such as sending emails, label management, and batch operations.
TypeScript
21.0K
4.5 points
M
Minimax MCP Server
The MiniMax Model Context Protocol (MCP) is an official server that supports interaction with powerful text-to-speech, video/image generation APIs, and is suitable for various client tools such as Claude Desktop and Cursor.
Python
47.5K
4.8 points
AIBase
Zhiqi Future, Your AI Solution Think Tank
© 2026AIBase