Cyberstrike
AI-powered offensive security agent with 7,300+ actionable security skills. Autonomous pentesting powered by MITRE ATT&CK (2,000+ Atomic tests), CIS Benchmarks (1,500+ controls), OWASP, NIST. Lazy-loading, zero context pollution. Your AI red team.
What is Cyberstrike?
Cyberstrike is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to ai-powered offensive security agent with 7,300+ actionable security skills. autonomous pentesting powered by mitre att&ck (2,000+ atomic tests), cis benchmarks (1,500+ controls), owasp, nist. lazy-loa...
AI-powered offensive security agent with 7,300+ actionable security skills. Autonomous pentesting powered by MITRE ATT&CK (2,000+ Atomic tests), CIS Benchmarks (1,500+ controls), OWASP, NIST. Lazy-loading, zero context pollution. Your AI red team.
This server falls under the Security category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- AI-powered offensive security agent with 7,300+ actionable s
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx cyberstrikeConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Cyberstrike
CyberStrike is an AI-powered offensive security agent that bundles 7,300+ actionable security skills grounded in MITRE ATT&CK atomic tests, CIS Benchmarks, OWASP, and NIST frameworks. It runs as a standalone terminal UI with a built-in proxy, shell execution, file operations, and web fetching — giving security teams a ready-made red team capability. Teams use it to run autonomous penetration tests across web applications, mobile apps, cloud environments, and internal networks without manually chaining individual tools.
Prerequisites
- Node.js 18+ (for npx installation) or Homebrew/Scoop for native install
- An LLM provider API key — Anthropic Claude, OpenAI, Google Gemini, or any of 15+ supported providers
- An MCP-compatible client such as Claude Desktop
- Optional: Cloudflare account if you need remote tunnel access to the web UI
Install CyberStrike
Install CyberStrike globally using npm, Homebrew (macOS), Scoop (Windows), or the shell installer. The npm route is the most portable.
npm i -g @cyberstrike-io/cyberstrike@latestLaunch and select your LLM provider
Run CyberStrike for the first time. The terminal UI will prompt you to select your LLM provider (Claude, GPT-4, Gemini, etc.) and supply your API key. No config files are required — setup is interactive.
npx cyberstrikeSet the web UI password (optional remote access)
If you want to expose the CyberStrike web interface via a Cloudflare Tunnel for remote use, set the server password environment variable. Local localhost connections bypass authentication automatically.
export CYBERSTRIKE_SERVER_PASSWORD=your-secure-passwordConfigure Claude Desktop (MCP mode)
Add CyberStrike to your Claude Desktop configuration to use it as an MCP server. The server exposes bash shell, file operations, web fetching, HTTP proxy interception, and vulnerability reporting tools.
{
"mcpServers": {
"cyberstrike": {
"command": "npx",
"args": ["cyberstrike"]
}
}
}Choose an agent specialization
CyberStrike ships with pre-built agent specializations: Web Application (OWASP WSTG), Mobile (Frida/Objection/MASTG), Cloud Security (AWS/Azure/GCP), Internal Network (Active Directory/Kerberos), and 8 proxy sub-testers. Select the appropriate agent for your engagement scope from the UI.
Run your first security assessment
Ask the agent to begin testing a target. It will autonomously plan and execute attack chains, collect evidence, and produce a vulnerability report. Always ensure you have explicit written permission before testing any target.
Cyberstrike Examples
Client configuration
Claude Desktop configuration for CyberStrike in MCP mode.
{
"mcpServers": {
"cyberstrike": {
"command": "npx",
"args": ["cyberstrike"]
}
}
}Prompts to try
Example prompts to use once CyberStrike is running as an MCP server.
- "Perform a web application penetration test on https://target.example.com using OWASP WSTG methodology"
- "Run MITRE ATT&CK atomic tests for credential dumping on this Linux host"
- "Assess the AWS environment for CIS Benchmark compliance and report all failing controls"
- "Intercept and replay the last HTTP request captured in the proxy session"
- "Generate a vulnerability report summarizing all findings from this engagement"Troubleshooting Cyberstrike
LLM provider API key not accepted on first launch
Ensure you are supplying the correct key for your chosen provider. For Anthropic Claude use an `sk-ant-...` key from console.anthropic.com. For OpenAI use an `sk-...` key. Re-run `npx cyberstrike` and re-enter the key when prompted.
Remote web UI access denied even with correct password
Confirm the `CYBERSTRIKE_SERVER_PASSWORD` environment variable is exported in the shell that starts CyberStrike. Remote connections (non-localhost) always require Basic Auth — the header format is `Authorization: Basic base64(username:password)`.
Shell tool executes commands but returns no output
Check that the process has the necessary permissions to run the underlying system tool (nmap, curl, etc.). Run `which <tool>` in a separate terminal to confirm the binary is on PATH and not blocked by a security policy.
Frequently Asked Questions about Cyberstrike
What is Cyberstrike?
Cyberstrike is a Model Context Protocol (MCP) server that ai-powered offensive security agent with 7,300+ actionable security skills. autonomous pentesting powered by mitre att&ck (2,000+ atomic tests), cis benchmarks (1,500+ controls), owasp, nist. lazy-loading, zero context pollution. your ai red team. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Cyberstrike?
Follow the installation instructions on the Cyberstrike GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Cyberstrike?
Cyberstrike works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Cyberstrike free to use?
Yes, Cyberstrike is open source and available under the AGPL-3.0 license. You can use it freely in both personal and commercial projects.
Cyberstrike Alternatives — Similar Security Servers
Looking for alternatives to Cyberstrike? Here are other popular security servers you can use with Claude, Cursor, and VS Code.
Casdoor
★ 13.6kAn open-source Agent-first Identity and Access Management (IAM) /LLM MCP & agent gateway and auth server with web UI supporting OpenClaw, MCP, OAuth, OIDC, SAML, CAS, LDAP, SCIM, WebAuthn, TOTP, MFA, Face ID, Google Workspace, Azure AD
ghidraMCP
★ 9.0kAn Model Context Protocol server that enables LLMs to autonomously reverse engineer applications by exposing Ghidra's decompilation and analysis tools. It allows AI agents to list code structures, rename methods, and analyze binaries directly through
HexStrike AI
★ 8.9kHexStrike AI MCP Agents is an advanced MCP server that lets AI agents (Claude, GPT, Copilot, etc.) autonomously run 150+ cybersecurity tools for automated pentesting, vulnerability discovery, bug bounty automation, and security research. Seamlessly b
IDA Pro MCP
★ 8.7kEnables AI-assisted reverse engineering in IDA Pro by providing tools to analyze binaries, decompile functions, manage comments, search patterns, and interact with the IDA database through natural language.
Anthropic Cybersecurity Skills
★ 6.6k754 structured cybersecurity skills for AI agents · Mapped to 5 frameworks: MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND & NIST AI RMF · agentskills.io standard · Works with Claude Code, GitHub Copilot, Codex CLI, Cursor, Gemini CLI & 20+ platform
Hooker
★ 5.1k🔥🔥 hooker is a Frida-based reverse engineering toolkit for Android. It offers a user-friendly CLI, universal scripts, auto hook generation, memory roaming to detect activities/services, one-click SOCKS5 proxy setup, Frida JustTrustMe, and BoringSSL u
Browse More Security MCP Servers
Explore all security servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Cyberstrike in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Cyberstrike?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.