Can it help with AI and LLM security?

Yes, it includes chapters on prompt injection, model poisoning, and RAG poisoning to help secure AI-driven applications and LLMs.

What are Ungar Games in this context?

Ungar Games are order-dependent security simulations that use join-semilattice structures to ensure attack chains respect logical prerequisites and state changes.

What is the BlackHat Go skill?

It is a specialized security capability for Claude Code that implements offensive and defensive techniques in Go, covering foundations from the 'Black Hat Go' book plus modern extensions like Web3 and AI.

What is AAIF integration?

AAIF (Agentic AI Foundation) integration allows the skill to work seamlessly with Model Context Protocol (MCP) and agent frameworks like goose for autonomous security tasks.

Does this skill support cloud and container security?

Yes, it includes specialized techniques for Kubernetes pod escapes, AWS SSRF, and container enumeration within the cloud security chapter.

Black Hat Go Security Knowledge Base

Name: Black Hat Go Security Knowledge Base
Author: plurigrid

byplurigrid

•

Security & Testing

Provides an extensive library of offensive security techniques and defensive patterns in Go for advanced red teaming and agentic simulations.

This skill transforms Claude into a security expert by integrating 186 techniques from the 'Black Hat Go' domain and extending them to modern frontiers like Cloud, Web3, and AI/ML security. It features a sophisticated knowledge base of 36 tools and 33 defenses, all structured within the AAIF (Agentic AI Foundation) ecosystem. Developers can leverage it for automated security audits, adversarial bisimulation, and multiplayer human-agent security exercises. By using MCP-native connectivity and order-dependent Ungar games, it provides a mathematically grounded framework for validating attack chains and defensive resilience across diverse environments from IoT to supply chain infrastructure.

Key Features

01Advanced coverage of Cloud (K8s/AWS), IoT, Web3, and GenAI security domains

02Adversarial bisimulation and Ungar games for rigorous attack chain validation

03Support for real-time multiplayer human-agent security exercises via CatColab

042 GitHub stars

05186 offensive techniques and 33 defenses across 37 specialized security chapters

06MCP-native integration for AAIF and goose agent workflows using AGENTS.md

Use Cases

01AI-agent-led security training and incident response simulations

02Validating LLM and GenAI security against OWASP Top 10 and prompt injection threats

03Automated red teaming and cloud infrastructure penetration testing

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add plurigrid/asi blackhat-go

For use in Claude.ai and ChatGPT

Download Skill