Security & Testing MCP Servers

Modular Server

Enables execution of various tasks on a server through a modular tool interface secured with path restrictions.

PRIMCS

Enables LLM agents to safely execute arbitrary Python code in a secure, throw-away sandbox.

Source

Enables AI assistants like Claude Desktop to directly interact with source code files and project directories for enhanced development workflows.

PHPCS

Provides PHP_CodeSniffer and PHP Code Beautifier and Fixer functionalities through a minimal Model Context Protocol server implemented in pure PHP.

Icecast

Analyzes Icecast streaming server configurations for security, performance, and capacity planning.

Amorce

Provides an open-source runtime enabling secure, cryptographically verifiable, and human-supervised transactions between AI agents, implementing the Amorce Agent Transaction Protocol (AATP).

Datalink

Provides secure database access for AI assistants to PostgreSQL, MySQL, and SQLite databases.

PuTTY

Exposes the full PuTTY toolchain as an MCP server, offering 30 Windows-native tools for SSH, SCP, SFTP, serial, key management, registry sessions, and multi-hop access.

Python Code Sandbox

Provides a secure, isolated environment for executing Python code in ephemeral Docker containers with automatic dependency management and file persistence.

Secure HTTP

Control and secure HTTP requests made by autonomous agents, ensuring full oversight of allowed actions and data.

Shrike

Protects AI agents from security threats like prompt injection, jailbreaks, and data exfiltration by scanning inputs and outputs.

Verified Repo Memory

Provides stale-proof repository memory for AI coding agents with just-in-time verification.

Epydios

Intercepts Model Context Protocol (MCP) tool calls, enforces access policies with approval workflows, records audit trails, and can proxy requests to downstream MCP servers.

Foundry zkSync

Enables AI assistants to interact with zkSync smart contracts and development tools through a comprehensive set of CLI commands.

PQVault

Manages post-quantum encrypted secrets for AI agents, offering secure access via a zero-knowledge proxy and the MCP protocol.

PolyCheck

Performs multi-language code analysis by combining AI-powered insights, static AST analysis, and Retrieval Augmented Generation to deliver intelligent code reviews.

T-IA Connect + AgentGateway

Securely enables AI agents to create, configure, and manage Siemens TIA Portal PLC projects with zero-trust security enforcement.

Vault

Manages and secures secrets for AI agents and missions, embedding OpenBao for robust cryptographic operations.

Aris

Unifies diverse AI models into a single operating system, intelligently routing tasks, managing memory, and automatically evolving for developers.

Redact

Automatically detects and obfuscates sensitive data in text and HTTP requests, preventing exposure to AI models during analysis.