chucknorris-mcp

A specialized MCP server providing LLM enhancement prompts and jailbreaks with dynamic schema adaptation, designed to augment AI capabilities via the MCP protocol.

• Category: ai-integration-mcp-servers
• Tags: mcp, llm-integration, prompt-management, open-source
• Source: GitHub Repository

Features

• MCP Server Implementation: Provides Model Context Protocol (MCP) server that delivers specialized enhancement prompts to LLMs.
• Jailbreak Prompts: Utilizes jailbreak prompts from elder-plinius' L1B3RT4S project.
• Dynamic Schema Adaptation: Implements a dynamic schema approach for prompt delivery, making jailbreak attempts less detectable by LLMs.
• Two-Phase Bypass: Uses a two-phase approach to bypass jailbreak detection (does not immediately present jailbreak content).
• Security Research Tool: Designed for security research and evaluation to identify vulnerabilities in LLM systems.
• Inspiration from Research: Implements techniques described in "The 'S' in MCP Stands for Security" and references Tool Poisoning Attack research by Invariant Labs.
• Compatibility: Intended for use with weaker models; effectiveness may vary with newer models (e.g., Claude, GPT-4).
• Open Source: Licensed under MIT.
• Community Supported: Part of the Pollinations.AI community.

Pricing

• Open Source: No pricing information; available under the MIT license.

Status

• Experimental: Work in progress with limitations; experimental effectiveness with newer LLMs.