damn-vulnerable-MCP-server

Name: damn-vulnerable-MCP-server
Author: harishsg993010

1.3kstars169forks

How to install damn-vulnerable-MCP-server?

harishsg993010/damn-vulnerable-mcp-server

$git clone https://github.com/harishsg993010/damn-vulnerable-MCP-server

Installs into the current project.

Summary

A deliberately vulnerable implementation of the Model Context Protocol (MCP) for educational purposes.

Files · 1

View on GitHub

README.md

1	# Damn Vulnerable Model Context Protocol (DVMCP)
2
3	A deliberately vulnerable implementation of the Model Context Protocol (MCP) for educational purposes.
4
5	## Overview
6
7	The Damn Vulnerable Model Context Protocol (DVMCP) is an educational project designed to demonstrate security vulnerabilities in MCP implementations. It contains 10 challenges of increasing difficulty that showcase different types of vulnerabilities and attack vectors.
8
9	This project is intended for security researchers, developers, and AI safety professionals to learn about potential security issues in MCP implementations and how to mitigate them.
10
11	## What is MCP?
12
13	The [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) is a standardized protocol that allows applications to provide context for Large Language Models (LLMs) in a structured way. It separates the concerns of providing context from the actual LLM interaction, enabling applications to expose resources, tools, and prompts to LLMs.
14
15	## Recommended MCP Clients
16
17	CLINE - VSCode Extension
18	Refer to this [Connecting to a Remote Server - Cline](https://docs.cline.bot/mcp/connecting-to-a-remote-server) for connecting Cline with MCP server
19
20	## Quick Start
21
22	Once you have cloned the repository, run the following commands:
23
24	```sh
25	docker build -t dvmcp .
26	docker run -p 9001-9010:9001-9010 dvmcp
27	```
28
29	## Disclaimer
30	It's not stable in a Windows environment. If you don't want to use Docker then please use Linux environment.
31	I recommend Docker to run the LAB and I am 100% percent sure it works well in the Docker environment
32
33	## Security Risks
34
35	While MCP provides many benefits, it also introduces new security considerations. This project demonstrates various vulnerabilities that can occur in MCP implementations, including:
36
37	1. Prompt Injection: Manipulating LLM behavior through malicious inputs
38	2. Tool Poisoning: Hiding malicious instructions in tool descriptions
39	3. Excessive Permissions: Exploiting overly permissive tool access
40	4. Rug Pull Attacks: Exploiting tool definition mutations
41	5. Tool Shadowing: Overriding legitimate tools with malicious ones
42	6. Indirect Prompt Injection: Injecting instructions through data sources
43	7. Token Theft: Exploiting insecure token storage
44	8. Malicious Code Execution: Executing arbitrary code through vulnerable tools
45	9. Remote Access Control: Gaining unauthorized system access
46	10. Multi-Vector Attacks: Combining multiple vulnerabilities
47
48	## Project Structure
49
50	```
51	damn-vulnerable-MCP-server/
52	├── README.md # Project overview
53	├── requirements.txt # Python dependencies
54	├── challenges/ # Challenge implementations
55	│ ├── easy/ # Easy difficulty challenges (1-3)
56	│ │ ├── challenge1/ # Basic Prompt Injection
57	│ │ ├── challenge2/ # Tool Poisoning
58	│ │ └── challenge3/ # Excessive Permission Scope
59	│ ├── medium/ # Medium difficulty challenges (4-7)
60	│ │ ├── challenge4/ # Rug Pull Attack
61	│ │ ├── challenge5/ # Tool Shadowing
62	│ │ ├── challenge6/ # Indirect Prompt Injection
63	│ │ └── challenge7/ # Token Theft
64	│ └── hard/ # Hard difficulty challenges (8-10)
65	│ ├── challenge8/ # Malicious Code Execution
66	│ ├── challenge9/ # Remote Access Control
67	│ └── challenge10/ # Multi-Vector Attack
68	├── docs/ # Documentation
69	│ ├── setup.md # Setup instructions
70	│ ├── challenges.md # Challenge descriptions
71	│ └── mcp_overview.md # MCP protocol overview
72	├── solutions/ # Solution guides
73	└── common/ # Shared code and utilities
74	```
75
76	## Getting Started
77
78	See the [Setup Guide](docs/setup.md) for detailed instructions on how to install and run the challenges.
79
80	## Challenges
81
82	The project includes 10 challenges across three difficulty levels:
83
84	### Easy Challenges
85
86	1. Basic Prompt Injection: Exploit unsaniti

People who liked this also viewed

6 picks

Type

odoo-mcp-proAI connector for Odoo ERP | Connect Claude, ChatGPT, Cursor and other AI tools to Odoo via MCPMCP ServersJul 2026551k63
omnigentUniversal autonomous agent framework with ReAct loop, multi-provider LLM routing, reasoning graph, and MCP integration, domain-agnostic for building specialized AI agents.MCP ServersFeb 202683k30
arxiv-mcp-serverAn MCP server that enables interacting with the arXiv API using natural languageMCP ServersAug 202576k40
United States WeatherProvide real-time and forecast weather information for locations in the United States using natural language queries. Access current conditions, multi-day and hourly forecasts, weather alerts, and nearby weather stations without requiring aMCP Serversrecently56k0
mcp-server-demoDemo of MCP Server in Python using FastMCPMCP ServersAug 202556k10
mcp-serverlearn to make some mcp serversMCP ServersMay 202556k5

damn-vulnerable-MCP-server

1.3kstars169forks

View on GitHub

How to install damn-vulnerable-MCP-server?

harishsg993010/damn-vulnerable-mcp-server

$git clone https://github.com/harishsg993010/damn-vulnerable-MCP-server

Installs into the current project.

Summary

A deliberately vulnerable implementation of the Model Context Protocol (MCP) for educational purposes.

Files · 1

View on GitHub

README.md

1	# Damn Vulnerable Model Context Protocol (DVMCP)
2
3	A deliberately vulnerable implementation of the Model Context Protocol (MCP) for educational purposes.
4
5	## Overview
6
7	The Damn Vulnerable Model Context Protocol (DVMCP) is an educational project designed to demonstrate security vulnerabilities in MCP implementations. It contains 10 challenges of increasing difficulty that showcase different types of vulnerabilities and attack vectors.
8
9	This project is intended for security researchers, developers, and AI safety professionals to learn about potential security issues in MCP implementations and how to mitigate them.
10
11	## What is MCP?
12
13	The [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) is a standardized protocol that allows applications to provide context for Large Language Models (LLMs) in a structured way. It separates the concerns of providing context from the actual LLM interaction, enabling applications to expose resources, tools, and prompts to LLMs.
14
15	## Recommended MCP Clients
16
17	CLINE - VSCode Extension
18	Refer to this [Connecting to a Remote Server - Cline](https://docs.cline.bot/mcp/connecting-to-a-remote-server) for connecting Cline with MCP server
19
20	## Quick Start
21
22	Once you have cloned the repository, run the following commands:
23
24	```sh
25	docker build -t dvmcp .
26	docker run -p 9001-9010:9001-9010 dvmcp
27	```
28
29	## Disclaimer
30	It's not stable in a Windows environment. If you don't want to use Docker then please use Linux environment.
31	I recommend Docker to run the LAB and I am 100% percent sure it works well in the Docker environment
32
33	## Security Risks
34
35	While MCP provides many benefits, it also introduces new security considerations. This project demonstrates various vulnerabilities that can occur in MCP implementations, including:
36
37	1. Prompt Injection: Manipulating LLM behavior through malicious inputs
38	2. Tool Poisoning: Hiding malicious instructions in tool descriptions
39	3. Excessive Permissions: Exploiting overly permissive tool access
40	4. Rug Pull Attacks: Exploiting tool definition mutations
41	5. Tool Shadowing: Overriding legitimate tools with malicious ones
42	6. Indirect Prompt Injection: Injecting instructions through data sources
43	7. Token Theft: Exploiting insecure token storage
44	8. Malicious Code Execution: Executing arbitrary code through vulnerable tools
45	9. Remote Access Control: Gaining unauthorized system access
46	10. Multi-Vector Attacks: Combining multiple vulnerabilities
47
48	## Project Structure
49
50	```
51	damn-vulnerable-MCP-server/
52	├── README.md # Project overview
53	├── requirements.txt # Python dependencies
54	├── challenges/ # Challenge implementations
55	│ ├── easy/ # Easy difficulty challenges (1-3)
56	│ │ ├── challenge1/ # Basic Prompt Injection
57	│ │ ├── challenge2/ # Tool Poisoning
58	│ │ └── challenge3/ # Excessive Permission Scope
59	│ ├── medium/ # Medium difficulty challenges (4-7)
60	│ │ ├── challenge4/ # Rug Pull Attack
61	│ │ ├── challenge5/ # Tool Shadowing
62	│ │ ├── challenge6/ # Indirect Prompt Injection
63	│ │ └── challenge7/ # Token Theft
64	│ └── hard/ # Hard difficulty challenges (8-10)
65	│ ├── challenge8/ # Malicious Code Execution
66	│ ├── challenge9/ # Remote Access Control
67	│ └── challenge10/ # Multi-Vector Attack
68	├── docs/ # Documentation
69	│ ├── setup.md # Setup instructions
70	│ ├── challenges.md # Challenge descriptions
71	│ └── mcp_overview.md # MCP protocol overview
72	├── solutions/ # Solution guides
73	└── common/ # Shared code and utilities
74	```
75
76	## Getting Started
77
78	See the [Setup Guide](docs/setup.md) for detailed instructions on how to install and run the challenges.
79
80	## Challenges
81
82	The project includes 10 challenges across three difficulty levels:
83
84	### Easy Challenges
85
86	1. Basic Prompt Injection: Exploit unsaniti

People who liked this also viewed

6 picks

Type