Skip to main content
mcpSource-backedReview first Safety Privacy

HexStrike AI MCP Server

Offensive security MCP framework that connects AI agents to a large toolkit for authorized penetration testing, vulnerability discovery, CTF, OSINT, and security research workflows.

by 0x4m4·added 2026-06-05·
Claude CodeClaude Desktop
HarnessClaude CodeClaude Desktop
Review first review before installing

Open the source and read safety notes before installing.

Safety notes

  • HexStrike AI orchestrates offensive security tools and can perform scanning, enumeration, exploitation support, password attacks, OSINT, and vulnerability research.
  • Use only on systems, domains, networks, binaries, accounts, and datasets where you have explicit authorization.
  • Agent-driven scans can create high traffic, trigger alarms, lock accounts, alter evidence, or disrupt services if scope and rate limits are not enforced.
  • Do not allow autonomous exploit generation, credential attacks, or destructive actions against third-party systems without human review and written permission.
  • Run in an isolated environment with controlled credentials, logging, allowlists, and clear stop conditions.

Privacy notes

  • Targets, scan results, credentials, tokens, exploit notes, screenshots, OSINT findings, and vulnerability reports may be sent to the MCP client and model.
  • Tool logs can contain customer data, secrets, infrastructure details, private IP ranges, bug bounty findings, and regulated incident information.
  • Review retention, telemetry, and sharing settings for every AI client and security tool connected to the workflow.

Prerequisites

  • Python 3.8 or newer.
  • Security tooling required for the selected modules, such as network, web, cloud, password, binary-analysis, OSINT, or CTF tools.
  • MCP client such as Claude Desktop, Cursor, VS Code Copilot, Roo Code, 5ire, or another compatible host.
  • Isolated lab, test network, bug bounty scope, or written authorization for every target.
  • Operator who understands penetration-testing law, safe-scope rules, and tool impact.

Schema details

Install type
cli
Troubleshooting
No
Source repository stats
Scope
Source repo
Collection metadata
Estimated setup
45 minutes
Difficulty
advanced
Tool listing metadata
Full copyable content
{
  "mcpServers": {
    "hexstrike-ai": {
      "command": "python3",
      "args": [
        "/ABSOLUTE_PATH_TO/hexstrike-ai/hexstrike_mcp.py",
        "--server",
        "LOCAL_HEXSTRIKE_SERVER_URL"
      ]
    }
  }
}

About this resource

Content

HexStrike AI MCP Server is an offensive-security automation framework for MCP clients. The project connects AI agents to a security-tool orchestration layer covering reconnaissance, web application testing, cloud security, password and authentication testing, reverse engineering, OSINT, CTF workflows, vulnerability intelligence, and reporting.

The README describes a multi-agent architecture with a decision engine, process management, caching, and a large catalog of security tools. It is intended for authorized penetration testing, bug bounty, CTF, and security research rather than general browser or coding automation.

Source Review

These sources were reviewed on 2026-06-05. Prefer the live repository for current client integration examples, supported security tools, server options, and platform-specific setup steps.

Features

  • MCP integration for Claude Desktop, Cursor, VS Code Copilot, Roo Code, 5ire, and other MCP-compatible agents.
  • Large security-tool catalog across network, web application, cloud, password, binary-analysis, CTF, and OSINT workflows.
  • Autonomous agent architecture for bug bounty, CTF, CVE intelligence, exploit generation support, and reporting workflows.
  • Local server process with Python MCP bridge configuration.
  • Visual reporting and vulnerability-card style output described by the project.

Installation

Clone the repository and install the Python dependencies:

git clone https://github.com/0x4m4/hexstrike-ai.git
cd hexstrike-ai
python3 -m venv hexstrike-env
pip3 install -r requirements.txt

Install only the security tools needed for your authorized workflow, then start the server:

python3 hexstrike_server.py

Configure your MCP client to run the repository's hexstrike_mcp.py bridge and provide the server argument documented in the README.

{
  "mcpServers": {
    "hexstrike-ai": {
      "command": "python3",
      "args": [
        "/ABSOLUTE_PATH_TO/hexstrike-ai/hexstrike_mcp.py",
        "--server",
        "LOCAL_HEXSTRIKE_SERVER_URL"
      ]
    }
  }
}

Use Cases

  • Run scoped reconnaissance and vulnerability discovery in a lab or bug bounty program.
  • Coordinate web application testing tools under a single MCP workflow.
  • Explore CTF challenges with binary-analysis, OSINT, and exploitation helpers.
  • Summarize authorized scan findings into triage notes and reports.
  • Compare tool results and ask an assistant to propose next safe test steps.

Safety and Privacy

HexStrike AI is powerful offensive-security infrastructure. Treat it as a penetration-testing workstation, not a general assistant plugin. Use explicit target allowlists, rate limits, isolated networks, disposable credentials, and human approval for exploitation, credential attacks, persistence, destructive actions, or reporting to third parties.

Assume prompts, logs, scan output, screenshots, exploit notes, and generated reports can contain sensitive or regulated security data. Keep findings in approved systems and remove secrets before sharing outputs with external model providers.

Duplicate Check

No 0x4m4/hexstrike-ai entry or source URL was found in content/mcp. This entry is separate from general security scanners and from reverse-engineering entries such as IDA Pro MCP and GhidraMCP.

#cybersecurity#penetration-testing#security#osint#ctf

Source citations

Signals

Loading live community signals…

More like this, weekly

A short, calm digest of reviewed Claude resources. Unsubscribe any time.