Vulnerable MCP Servers Lab

MCPOfficialOpen SourceMIT73.3

by appsecco • Uncategorized

A collection of intentionally vulnerable MCP server implementations for security training and research.

Example Use Cases

1
Demonstrate and research common MCP server vulnerabilities.
2
Hands-on training in exploiting MCP server security flaws.
3
Understand risks of integrating untrusted tools and content into AI workflows.

Description

This repository provides multiple vulnerable MCP servers, each demonstrating different security weaknesses such as path traversal, code execution, prompt injection, supply-chain compromise, and secrets exposure. It is designed for hands-on demos and research to help pentesters and AI Red Teamers understand risks in MCP server integrations. The servers come with detailed READMEs explaining their functionality, how to run them, and how to exploit their vulnerabilities in a controlled lab environment.

Capabilities(2 total)

Tools (2)

get_account

Get basic profile details for a user by username (id, name, handle, bio, follower/following counts, created_at).

username:string*

get_tweets

Get recent tweets for a user. Inputs: username (required), limit (optional, default 10, max 50). Returns tweet id, text, created_at.

username:string*limit:number

Quick Actions

View on GitHub

Security

Scanned 1 month(s) ago

Risk Level

MINIMAL

Read-only data retrieval, no side effects

Trust Score

C60/100

6/17 checks passed

Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.

Quick Stats

Service TypeMCP

Pricing ModelFree

Capabilities2 Tools / 0 Prompts / 0 Resources

Ownerappsecco

CategoryUncategorized

DependenciesStandalone

Set Your Username