j3ssie/osmedeus
Osmedeus is a powerful and modern orchestration engine designed for reconnaissance and penetration testing workflows, written in Go.
Awesome AI Agents › Red-Teaming Platforms
ARCANGEL0/EVA
EVA, short for Exploit Vector Agent, is a terminal application that puts a large language model in the loop of an offensive security engagement. The operator starts a session, picks a model backend and defines a target; EVA then proposes an attack strategy, generates concrete shell commands for the current phase, reads the output of each command and uses it to decide what to try next. The documented flow moves through reconnaissance and enumeration, vulnerability assessment with CVE lookup, exploitation planning and execution, privilege escalation and post-exploitation, and ends with a generated report; when a step fails the agent loops back with alternative approaches instead of stopping. Sessions are persistent, stored under a local data directory, so an engagement can be paused and resumed with its chat history, findings and mapped network state intact, and results including network graphs can be exported. Model access is deliberately pluggable: Ollama for fully local inference, OpenAI, Anthropic and Google Gemini for hosted models, G4F providers, or any custom OpenAI-compatible endpoint, all configured through an eva --config flow that also handles API keys, model download and dependency setup. Installation is a pip package or a direct clone of the single eva.py entry point. The stated goal is assistance rather than replacement: EVA is meant to speed up a professional tester by handling the mechanical parts of enumeration and by suggesting paths a human might not consider, while the human keeps control over what is actually executed. It is aimed at penetration testers, red teamers and CTF players working on Linux, and, like any tool that runs generated commands against live hosts, it requires authorisation for the target and careful review of what the model proposes.
https://github.com/ARCANGEL0/EVA
Also in Red-Teaming Platforms
Tencent/AI-Infra-Guard
AI-Infra-Guard by Tencent is a comprehensive AI Red Teaming platform offering intelligent vulnerability scanning, MCP server risk detection, and jailbreak evaluation to enhance AI security.
beclab/Olares
Olares is an open-source sovereign cloud operating system that enables users to run local AI models, manage personal data, and self-host various applications securely and privately on their own hardware.
GH05TCREW/pentestagent
Python terminal agent for AI-assisted penetration testing that drives nmap, sqlmap, a browser and MCP servers, with assist, autonomous agent, multi-agent crew and guided interactive modes.
larlarua/AutoCVE
Agent-driven platform that automates CVE discovery end to end, using multiple coordinated agents to audit source code, verify vulnerabilities, and generate submission-ready reports.
SanMuzZzZz/LuaN1aoAgent
LuaN1aoAgent v2 is an autonomous security research agent that splits work between Planner, Executor and Observer roles and records every conclusion as evidence-backed nodes in a causal reasoning graph.
qualifire-dev/rogue
Client-server platform for evaluating and red-teaming AI agents over A2A, MCP or direct Python calls, with 75+ vulnerabilities, 20 attack techniques and compliance-framework reporting.
ASCIT31/Dark-Moon
Open-source autonomous penetration testing platform where specialised agents plan and run real offensive operations through a controlled MCP execution layer, with local tokenization that hides sensitive values from the model.