LibreAI / Agents IA / hexstrike-ai
// agents · 0x4m4/hexstrike-ai

hexstrike-ai

hexstrike-ai est un projet de agents IA open source, créé par 0x4m4 (0x4m4/hexstrike-ai sur GitHub). Il compte 12 532 étoiles sur GitHub et 2 549 forks, il est publié sous la licence MIT et a été mis à jour pour la dernière fois sur GitHub le 3 août 2026.

MCP server letting AI agents run 150+ cybersecurity tools for authorized pentesting and research.

★ 12,5 k étoiles⑂ 2,5 k forksMITmis à jour août 2026
▦ visuel — le README de ce projet ne contient aucune capture intégrable,
voici donc un emplacement réservé. Le guide ci-dessous est complet malgré tout.
// détails du projet

Détails du projet

Étoiles GitHub
12 532
Forks
2 549
Licence
MIT
Dernière mise à jour sur GitHub
3 août 2026
Catégorie
Agents IA
Type de fiche
Guide d'installation complet
Site officiel
Official site
// fonctionnement

Comment ça marche

  1. Confirm authorization and isolation

    Only test systems you own or have written authorization for, and plan to run HexStrike in a dedicated security-testing VM, as the README advises.

  2. Set up the environment

    Clone the repo on a Kali Linux-style machine, create a virtualenv, install the Python requirements, and make sure the security tools you need are installed.

  3. Start the server

    Run hexstrike_server.py (add --debug for detailed logs) and verify it is listening on port 8888.

  4. Connect your AI agent and scope a task

    Point your MCP client at hexstrike_mcp.py, then describe an authorized assessment - for example network scanning or web application testing on your own site - and monitor activity in the dashboard.

// fonctionnalités

Pourquoi on l'adore

150+ integrated security tools

One MCP interface fronts a comprehensive arsenal for network scanning, web testing, vulnerability assessment and more, across 35+ attack categories.

12+ autonomous AI agents

Specialized agents make decisions and manage workflows, chaining tools together for complete assessments.

Thousands of vulnerability templates

Nuclei integration with 4000+ templates gives broad, current vulnerability coverage.

Real-time monitoring

A live process dashboard (via the server's API) lets you watch, inspect and terminate agent-run processes.

Process and error management

The server tracks running processes and includes debug modes and troubleshooting tooling for MCP connections and missing binaries.

Built for Kali-style environments

Designed around the standard penetration-testing toolset, with checks such as which nmap/gobuster/nuclei to verify tool availability.

Advertisement
// lancez-le

Lancez-le vous-même

  1. Clone the repository

    Clone HexStrike AI and enter the project directory.

    git clone https://github.com/0x4m4/hexstrike-ai.git
    cd hexstrike-ai
  2. Install dependencies

    Create a virtual environment and install the Python requirements; install the underlying security tools (nmap, nuclei, etc.) from their official sources.

    python3 -m venv hexstrike-dev
    source hexstrike-dev/bin/activate
    pip install -r requirements.txt
  3. Start the server

    Launch the HexStrike server, then configure your MCP client to run hexstrike_mcp.py so your agent can call the tools.

    python3 hexstrike_server.py --port 8888 --debug
// cas d'usage

Ce que vous pouvez créer

Advertisement
// faq

Questions fréquentes

Who maintains HexStrike AI?

It was created by 0x4m4 and is owned and developed by OTT Cybersecurity LLC, with community contributions.

Is it free to use?

Yes, it is MIT-licensed. Legal and ethical use is limited to authorized testing, bug bounties within scope, CTFs and research on systems you own - the README is explicit about this.

Does it need a paid API key?

The server itself does not; the AI agent you connect (Claude, GPT, Copilot, etc.) uses your own subscription or API key.

How does HexStrike compare with PentestGPT?

PentestGPT is an agent framework that drives general coding-agent backends through a staged pipeline; HexStrike is an MCP tool server that exposes 150+ installed security tools to whichever MCP-capable agent you already use.

How do I get started?

On a Kali-style VM, clone the repo, pip install -r requirements.txt, run python3 hexstrike_server.py, and connect your MCP client via hexstrike_mcp.py.

// continuez l'exploration

Outils similaires

Agents IA

apify-mcp-server

MCP server that lets AI agents run thousands of Apify Store scrapers and crawlers as tools.

★ 10,1 k étoilesMIT
Voir le guide →
Agents IA

MobileBuildMCP

An MCP server and CLI that lets AI agents build, test and run iOS and macOS projects.

★ 6,5 k étoilesMIT
Voir le guide →
Agents IA

deep-research

Deep Research with any LLM: in-depth reports in minutes, with an SSE API and MCP server built in

★ 4,7 k étoilesMIT
Voir le guide →
Agents IA

excel-mcp-server

Excel MCP Server: let AI agents read, write, format, and chart real .xlsx workbooks — no Excel required

★ 4,2 k étoilesMIT
Voir le guide →
Agents IA

brave-search-mcp-server

Brave's official MCP server: web, news, image, video, local, and LLM-context search for your agents.

★ 1,5 k étoilesMIT
Voir le guide →
Agents IA

chrome-devtools-mcp

Chrome DevTools for coding agents: an MCP server that lets agents debug and inspect real Chrome

★ 53,1 k étoilesApache-2.0
Voir le guide →
// alternatives

Alternatives en Agents IA

Les autres projets de la catégorie Agents IA avec le plus d'étoiles du catalogue LibreAI — la sélection réaliste d'alternatives si hexstrike-ai ne vous convient pas.

// plus en agents

Plus en Agents IA

Projets de la catégorie Agents IA proches dans le catalogue, les plus proches en étoiles de hexstrike-ai — continuez à explorer de part et d'autre.

Obtenir hexstrike-ai

Gratuit et ouvert : récupérez le code source, mettez une étoile au dépôt et commencez à créer.