strawman

Red-team text prompts for vulnerabilities using adversarial inputs.

6|1|Updated Jun 23, 2026

Search Tags:#security #prompt engineering #LLM #red-teaming #prompt hardening

Ready to run

npx skills add https://github.com/JGalego/Bag-of-Tricks --skill strawman

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: strawman
Source: https://github.com/JGalego/Bag-of-Tricks/tree/main/strawman
Command: npx skills add https://github.com/JGalego/Bag-of-Tricks --skill strawman

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires anthropic, openai, google-genai, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enables you to red-team prompts before shipping them, ensuring they are robust against adversarial inputs and potential security breaches.

Core Features & Use Cases

Adversarial Attack Simulation: Simulates various adversarial attacks on prompts, including jailbreaks, instruction injection, scope derailment, and more.
Prompt Hardening: Provides detailed feedback on where prompts have weaknesses and how to harden them.
Use Case: Before deploying a system prompt for an LLM, use this Skill to identify any potential vulnerabilities and mitigate them.

Quick Start

Run the strawman skill on your system prompt with the following command: strawman my_system_prompt.txt

Frequently Asked Questions about strawman

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I red-team a prompt to test for jailbreaks before deployment?▼

To red-team a prompt for jailbreaks, provide your system prompt in a text file. The skill simulates adversarial attacks like jailbreaks and instruction injection, identifies scope enforcement failures, and provides feedback on weaknesses with suggested fixes for prompt hardening.

What is prompt hardening and how does it protect LLM applications?▼

Prompt hardening strengthens system prompts against adversarial inputs. It identifies vulnerabilities like scope derailment and security breaches, then applies suggested fixes to ensure the LLM withstands malicious attacks before deployment.

Do I need API access to OpenAI or Anthropic to test prompt vulnerabilities?▼

Yes, testing prompt vulnerabilities requires API access to LLM backends like Anthropic, OpenAI, or Gemini. The red-teaming process uses these APIs to simulate adversarial attacks and evaluate your system prompt's responses to malicious inputs.

How do I check if my system prompt is vulnerable to instruction injection?▼

To check for instruction injection vulnerabilities, run an adversarial attack simulation against your system prompt. This identifies security breaches and scope enforcement failures, providing detailed feedback on weaknesses and hardening recommendations.

What types of adversarial attacks should I test my LLM prompts against before shipping?▼

Before shipping LLM prompts, test against adversarial attacks including jailbreaks, instruction injection, and scope derailment. Simulating these attacks identifies security breaches and scope enforcement failures, ensuring your prompt withstands malicious inputs.

Can I use this prompt red-teaming skill with Google Gemini models?▼

Yes, this skill supports Google Gemini models. It requires API access to LLM backends like Anthropic, OpenAI, or Google Gemini to simulate adversarial attacks and provide feedback on system prompt vulnerabilities and suggested fixes.