官方说明2 个来源MCP 服务2 天前
noor202401938-netizen/cyber-chef-mcp
CyberChef Model Context Protocol (MCP) Server data transformations, ciphers, hashing, JWT decoding, and payload deobfuscation for AI security & coding agents.
在 GitHub 查看 ↗直链 · 无中转页
配置模板（请按你的客户端调整）
claude mcp add cyber-chef-mcp -- npx -y cyber-chef-mcp
这是根据仓库名生成的可粘贴骨架，命令行参数与包名请以该项目 README 为准。
- ★ 数取值
- 1 ★ · noor202401938-netizen/cyber-chef-mcp · 取自 GitHub 搜索 API · （2026-10-16 20:00:20 核对）
- 来源明细
- 收录自 punkpeye/awesome-mcp-servers 榜单（我们从中收集的条目之一） · 条目类型：MCP 服务
- 收录与活跃
- 收录于 2026-10-15 · 最近见到 2026-10-16 · 共见到 2 次 · 被 2 个来源列出
- 评分构成
- 综合分 203 · 1 ★ · 2 个来源 · mcp
常见问题（答案都来自我们的记录）
怎么安装？
我们【没有】从该仓库取到安装命令（我们的 install 覆盖率只有约 4%）。请看它的 README 或 releases 页。
它适合谁 / 用来做什么？
我们把它归在这些标签下: MCP 服务。标签来自条目自身，不是我们强加的分类。
谁维护它、还在更新吗？
仓库
noor202401938-netizen/cyber-chef-mcp；我们收录于 2026-10-15；最近见到 2026-10-16。这里没有"官方认证"或"质量评级"——我们只记录来源与时间。第三方质量标记
🍳 CyberChef MCP Server
以下摘录自该仓库的 README（片段） · 在 GitHub 读完整 README →
Deterministic Cryptographic Engine & Multi-Stage Deobfuscation Layer for AI Agents
-purple.svg?style=flat-square)
**A pure-JavaScript Model Context Protocol (MCP) server that empowers autonomous AI agents (Claude Code, Cursor, Windsurf, Codex, Strix) to execute complex multi-stage deobfuscation recipes, evaluate calibrated Shannon entropy, parse modern password hashes (Argon2/bcrypt), and scan for PII/DLP data without LLM hallucination.**
🌐 Live Cloud Dashboard • ⚡ Remote SSE Endpoint • 📦 NPM Package • 📜 MCP Server Card • 🤖 llms.txt
---
🌐 Live Cloud Deployment (Azure App Service)
CyberChef MCP is continuously hosted on Microsoft Azure with full HTTPS, SSE streaming, and discovery endpoints active:
| Endpoint | Method | URL | Description |
|:---|:---:|:---|:---|
| **Deobfuscation Sandbox** | GET | / | Interactive web dashboard, live in-browser recipe execution & agent configuration hub |
| **Remote MCP SSE** | GET | /sse | Remote Server-Sent Events MCP endpoint for cloud-connected AI agents |
| **JSON-RPC Messages** | POST | /messages | Bidirectional MCP JSON-RPC execution gateway |
| **Health Probe** | GET | /health | Real-time liveness check reporting server status and tool count (17 tools, 28 operations) |
| **MCP Server Card** | GET | /.well-known/mcp/server-card.json | Standardized Model Context Protocol discovery catalog and schema |
| **Agent llms.txt** | GET | /llms.txt | Compact machine-readable reference optimized for LLM reasoning ingestion |
---
⚡ Token Economics: Why Use CyberChef MCP?
Instead of burning thousands of output tokens having an LLM write, debug, and execute Python scripts in a sandbox, CyberChef MCP provides **instantaneous, deterministic execution** in a single tool call:
| Task | Standard LLM (Python Execution) | CyberChef MCP Tool Call | Token Savings | Latency Speedup |
|---|---|---|---|---|
| **Multi-layer Deobfuscation (Hex → XOR → B64)** | 3 turns, ~1,850 tokens, 4,200ms | **1 turn, ~28 tokens, 0.04ms** | **98.5% fewer tokens** | **105,000x faster** |
| **Shannon Entropy Calculation** | 2 turns, ~920 tokens, 2,100ms | **1 turn, ~18 tokens, 0.016ms** | **98.0% fewer tokens** | **131,250x faster** |
| **JWT Decode & Expiry Check** | 2 turns, ~750 tokens, 1,800ms | **1 turn, ~22 tokens, 0.022ms** | **97.1% fewer tokens** | **81,800x faster** |
---
🚀 Quick Start & Integration
Option 1: Remote Azure SSE (Zero Setup Needed)
Connect your AI agent directly to the live cloud endpoint without installing any local packages:
In Claude Desktop (claude_desktop_config.json)
In Cursor (~/.cursor/mcp.json)
---
Option 2: Local Stdio via NPX (Recommended for Local Privacy)
Run locally on your workstation for zero network latency, air-gapped security, and offline operation:
Configuration (claude_desktop_config.json or mcp.json)
Via Claude Code CLI
Via Smithery (1-Click Install)
---
🛠️ MCP Tools Catalog (17 Primary Forensic & Crypto Operations)
| Tool Name | Parameters | Forensic Purpose & Description |
|:---|:---|:---|
| **cyberchef_bake** | input: string recipe: array | **Multi-Stage Recipe Pipeline:** Execute complex sequential operations across 28+ core decoders (Hex $\to$ XOR $\to$ Base64 $\to$ Gunzip) with timeout and memory safety guards. |
| **cyberchef_magic** | input: string | **Automated Heuristic Decoder:** Automatically identifies unknown encodings, single-byte XOR keys, and compressed data streams without prior knowledge. |
-purple.svg?style=flat-square)
**A pure-JavaScript Model Context Protocol (MCP) server that empowers autonomous AI agents (Claude Code, Cursor, Windsurf, Codex, Strix) to execute complex multi-stage deobfuscation recipes, evaluate calibrated Shannon entropy, parse modern password hashes (Argon2/bcrypt), and scan for PII/DLP data without LLM hallucination.**
🌐 Live Cloud Dashboard • ⚡ Remote SSE Endpoint • 📦 NPM Package • 📜 MCP Server Card • 🤖 llms.txt
---
🌐 Live Cloud Deployment (Azure App Service)
CyberChef MCP is continuously hosted on Microsoft Azure with full HTTPS, SSE streaming, and discovery endpoints active:
| Endpoint | Method | URL | Description |
|:---|:---:|:---|:---|
| **Deobfuscation Sandbox** | GET | / | Interactive web dashboard, live in-browser recipe execution & agent configuration hub |
| **Remote MCP SSE** | GET | /sse | Remote Server-Sent Events MCP endpoint for cloud-connected AI agents |
| **JSON-RPC Messages** | POST | /messages | Bidirectional MCP JSON-RPC execution gateway |
| **Health Probe** | GET | /health | Real-time liveness check reporting server status and tool count (17 tools, 28 operations) |
| **MCP Server Card** | GET | /.well-known/mcp/server-card.json | Standardized Model Context Protocol discovery catalog and schema |
| **Agent llms.txt** | GET | /llms.txt | Compact machine-readable reference optimized for LLM reasoning ingestion |
---
⚡ Token Economics: Why Use CyberChef MCP?
Instead of burning thousands of output tokens having an LLM write, debug, and execute Python scripts in a sandbox, CyberChef MCP provides **instantaneous, deterministic execution** in a single tool call:
| Task | Standard LLM (Python Execution) | CyberChef MCP Tool Call | Token Savings | Latency Speedup |
|---|---|---|---|---|
| **Multi-layer Deobfuscation (Hex → XOR → B64)** | 3 turns, ~1,850 tokens, 4,200ms | **1 turn, ~28 tokens, 0.04ms** | **98.5% fewer tokens** | **105,000x faster** |
| **Shannon Entropy Calculation** | 2 turns, ~920 tokens, 2,100ms | **1 turn, ~18 tokens, 0.016ms** | **98.0% fewer tokens** | **131,250x faster** |
| **JWT Decode & Expiry Check** | 2 turns, ~750 tokens, 1,800ms | **1 turn, ~22 tokens, 0.022ms** | **97.1% fewer tokens** | **81,800x faster** |
---
🚀 Quick Start & Integration
Option 1: Remote Azure SSE (Zero Setup Needed)
Connect your AI agent directly to the live cloud endpoint without installing any local packages:
In Claude Desktop (claude_desktop_config.json)
In Cursor (~/.cursor/mcp.json)
---
Option 2: Local Stdio via NPX (Recommended for Local Privacy)
Run locally on your workstation for zero network latency, air-gapped security, and offline operation:
Configuration (claude_desktop_config.json or mcp.json)
Via Claude Code CLI
Via Smithery (1-Click Install)
---
🛠️ MCP Tools Catalog (17 Primary Forensic & Crypto Operations)
| Tool Name | Parameters | Forensic Purpose & Description |
|:---|:---|:---|
| **cyberchef_bake** | input: string recipe: array | **Multi-Stage Recipe Pipeline:** Execute complex sequential operations across 28+ core decoders (Hex $\to$ XOR $\to$ Base64 $\to$ Gunzip) with timeout and memory safety guards. |
| **cyberchef_magic** | input: string | **Automated Heuristic Decoder:** Automatically identifies unknown encodings, single-byte XOR keys, and compressed data streams without prior knowledge. |
- 来源
- awesome · https://github.com/noor202401938-netizen/cyber-chef-mcp
- 按领域浏览
- MCP 服务
- Traction
- 1 ★
- 收录于
- 昨天
更多MCP 服务
所属领域与标签
每个标签都能点进去看同类的全部条目。数字是当前库里的真实条数。