A military-grade adversarial analysis engine that deploys 32 parallel agents to stress-test content and validate ideas through rigorous competition.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install tesy
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install tesy using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is RedTeam?
The RedTeam skill is a powerhouse for adversarial analysis, specifically designed for users of Openclaw Skills who need to stress-test high-stakes decisions, code architectures, or business plans. By deploying a fleet of 32 parallel agents—ranging from senior engineers to professional pentesters—it provides a level of scrutiny that goes far beyond standard AI interactions.
Whether you are attacking a complex microservices proposal or looking for the single point of failure in a business strategy, RedTeam provides the feedback necessary to harden your logic. It utilizes a sophisticated Five-Phase Protocol to ensure that every critique is backed by convergent insights, high-quality steelman representations, and devastating counter-arguments.
RedTeam Use Cases
- Stress-testing infrastructure migration plans and software architecture proposals.
- Identifying hidden vulnerabilities in business strategies and financial models.
- Performing adversarial validation between competing content or feature proposals.
- Generating high-fidelity steelman arguments to understand the strongest version of an opposing view.
- Conducting bot battles to synthesize the best possible solution from multiple AI perspectives using Openclaw Skills.
How RedTeam Works
- Decomposition: The skill breaks the user input into 24 atomic claims for granular, high-resolution analysis.
- Parallel Analysis: 32 specialized agents examine these claims simultaneously from diverse perspectives, identifying both hidden strengths and critical weaknesses.
- Synthesis: The engine identifies convergent insights where multiple expert agents agree on a specific risk or opportunity.
- Steelman: It constructs the strongest possible version of the original argument to ensure the critique is fair and rigorous.
- Counter-Argument: It produces an 8-point devastating rebuttal based on the parallel analysis findings to expose the core issues.
RedTeam Setup
To integrate RedTeam into your environment, follow these configuration steps for Openclaw Skills:
# Optional: Create a customization directory to override default behaviors
mkdir -p ~/.claude/skills/CORE/USER/SKILLCUSTOMIZATIONS/RedTeam/
# Place any specific preferences or resources in this directory
touch ~/.claude/skills/CORE/USER/SKILLCUSTOMIZATIONS/RedTeam/PREFERENCES.md
The skill will automatically detect this directory and apply your custom rules during execution.
RedTeam Data Schema & Taxonomy
The RedTeam skill generates a structured report within the Openclaw Skills ecosystem as detailed below:
| Component | Type | Description |
|---|---|---|
| Atomic Claims | List | A 24-point breakdown of the subject matter |
| Parallel Insights | Array | Convergent findings from 32 expert agent perspectives |
| Steelman | Markdown | The strongest defensible version of the input |
| Counter-Arguments | List | 8 critical points of failure or rebuttals |
| Voice Notifications | Webhook | Real-time execution status via local API |
RedTeam Advanced Features
- Multi-agent orchestration involving 32 distinct personas for comprehensive coverage.
- Five-Phase Protocol ensures high-fidelity results from decomposition to final rebuttal.
- Local customization support allowing users to override defaults with specific project context.
- Integrated voice notification system via curl for real-time workflow tracking.
- Seamless routing between Parallel Analysis and Adversarial Validation workflows depending on user intent within Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
Yao Liu Yao Divination
A digital implementation of the traditional Chinese Liu Yao divination method for generating and interpreting hexagrams.
Bazi Chart Generator
A specialized tool for generating traditional Chinese Bazi charts and 10-step Great Luck cycles based on birth data.
Security Audit Skill
A comprehensive security auditing toolkit for scanning codebases, infrastructure, and agentic AI systems for vulnerabilities and misconfigurations.
Response Speed Test
A high-precision performance monitoring tool that provides millisecond-level measurement of the AI agent response lifecycle.
CloudMonkey
A comprehensive interface for managing Apache CloudStack infrastructure via the CloudMonkey (cmk) CLI tool.
Clawpay
A payment delivery skill enabling AI agents and humans to exchange value via secure crypto payment requests and result delivery.
Featured*
