red-team-assessment

Conduct adversarial security assessments and generate remediation plans with multiple LLMs.

10|Updated Mar 31, 2026

Search Tags:#LLM #AI analysis #red team #MCP tools #security assessment

Ready to run

npx skills add https://github.com/verivus-oss/llm-cli-gateway --skill red-team-assessment-verivus-oss

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: red-team-assessment
Source: https://github.com/verivus-oss/llm-cli-gateway/tree/main/skills/red-team-assessment
Command: npx skills add https://github.com/verivus-oss/llm-cli-gateway --skill red-team-assessment-verivus-oss

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enables red team security assessments using AI for code, architecture, or configurations, ensuring thorough analysis with approvals and optional tools like sqry, exa, and ref_tools.

Core Features & Use Cases

Red Team Security Assessment: Submit code, designs, or configurations to LLMs for adversarial security analysis.
LLM Dispatch Defaults: Configure dispatch defaults for model usage, approval strategies, and timeouts.
Multi-LLM Red Team: Run assessments with multiple LLMs (Claude, Codex, Gemini, Grok, Mistral) for comprehensive coverage.
Blue Team Response: Generate defensive remediation plans for identified vulnerabilities with different LLMs.

Quick Start

Perform a red team assessment on the file 'security-review.docx' using the Claude LLM.

Frequently Asked Questions about red-team-assessment

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

FAQPage Schema

How do I conduct a red team security assessment using AI?▼

You can perform a red team security assessment by submitting code, architecture designs, or configurations to LLMs for adversarial analysis. The Skill handles assessment requests and generates defensive remediation plans using managed approval processes and MCP tools.

Can I use multiple LLMs for comprehensive security analysis?▼

Yes, you can run a multi-LLM red team assessment with Claude, Codex, Gemini, Grok, and Mistral for comprehensive coverage. Using multiple models provides thorough adversarial analysis and supports generating blue team remediation plans.

What MCP tools are needed for LLM-based security assessments?▼

LLM-based red team assessments optionally use MCP tools like sqry, exa, and ref_tools for comprehensive analysis. These tools support the assessment process by enabling managed approval processes and tool invocation during adversarial analysis.

How do I generate blue team remediation plans for identified vulnerabilities?▼

To generate a blue team response, the Skill processes identified vulnerabilities from the red team assessment and creates defensive remediation plans. You can use different LLMs to generate these defensive plans for comprehensive coverage.

Do I need to configure dispatch defaults before running AI red teaming?▼

Yes, you can configure LLM dispatch defaults for model usage, approval strategies, and timeouts before running AI red teaming. These settings manage how the assessment handles adversarial analysis requests and tool invocation.

Does this AI analysis tool support managed approval processes for security reviews?▼

Yes, this AI analysis tool supports managed approval processes and tool invocation during security reviews. It requires access to multiple LLMs and MCP tools to handle adversarial analysis requests and generate remediation plans.