M507/RamiGPT

Autonomous Privilege Escalation using AI + Benchmarking models

What it solves

RamiGPT is an AI-powered offensive security agent designed to automate the process of privilege escalation. It helps security researchers and penetration testers find a path to root access on a target system by autonomously navigating misconfigurations and vulnerabilities.

How it works

The system operates in a "Full AI" loop: it analyzes the current session context, generates a shell command via a configured LLM, executes that command over SSH, and feeds the output back into the model's history. This cycle repeats until root access is achieved or a request budget is exhausted. It integrates with PwnTools and provides a web-based workspace for managing multiple SSH sessions, guiding the model with hints, and running enumeration tools like LinPEAS.

Who it’s for

It is designed for offensive security professionals, penetration testers, and researchers who want to automate the discovery of privilege escalation paths on Linux systems.

Highlights

  • Autonomous Priv-Esc Loop: A self-correcting loop that attempts to gain root access without manual intervention.
  • Multi-Backend Support: Compatible with Ollama, Open WebUI, OpenAI, OpenRouter, and Cursor API.
  • Multi-Session Workspace: A web interface for managing multiple SSH targets with a built-in terminal and AI control panel.
  • Integrated Benchmarking: A comprehensive suite of 285 intentionally misconfigured targets to evaluate the performance and token efficiency of different LLMs in security tasks.

Related

  • Project
    ipa-lab/hackingBuddyGPT

    An open-source framework for building LLM-driven security-testing agents that automates tasks like privilege escalation and web pentesting.

  • Project
    HolmesGPT/holmesgpt

    An open-source AI agent for SREs that automates production incident investigation and root cause analysis across any infrastructure stack.

  • Project
    zylon-ai/private-gpt

    An open-source API layer that turns local LLM inference servers into production-ready AI applications by providing RAG, tool use, and Claude API compatibility.

  • Project
    yashab-cyber/HackGpt

    An AI-powered penetration testing platform for enterprise security teams that automates vulnerability assessments and compliance mapping using a multi-provider AI engine.

  • Project
    bugbasesecurity/pentest-copilot

    An AI-driven penetration testing agent that autonomously runs security tools on a Kali attack box, analyzes results, and iterates to find vulnerabilities.