Havuz

Global edition of ARTEX — an LLM multi-agent autonomous penetration-testing framework. English default, switchable to Korean/Chinese/Spanish

A professional, high-fidelity de-anonymization and vulnerability diagnostic framework. Features a zero-coupling modular architecture for bro

Cold Coffee, model görev şablonları oluşturma, dosya paketlerini yönetme ve OpenAI-compatible API üzerinden görev gönderme işlevlerini bir masaüstü çalışma alanında birleştiriyor.

Neden yükseliyor · 41 günde 723 yıldız ve son 28 günde 32 commit dikkat çekiyor; ancak commitlerin tamamının tek yazardan gelmesi ve README'nin gerçek model başarısını henüz doğrulanmamış sayması, bu ilginin ürünün etkinliğine dair kanıt olmadığını gösteriyor.

open·kritt, AI agents kullanarak kod depolarındaki olası güvenlik açıklarını tarayan, doğrulamaya ve önceliklendirmeye yönelik self-hosted bir platformdur.

Neden yükseliyor · 69 günde 2171 yıldız alması dikkat çekici erken ilgiye işaret ediyor; son 28 günde 27 commit ve 5 farklı yazar da projenin aktif geliştirildiğini gösteriyor. Ancak yıldızlar ve README'deki iddialar, taramaların gerçek açık bulma başarısını veya yanlış pozitif oranını kanıtlamıyor.

0, tarayıcı, terminal ve MCP üzerinden güvenlik iş akışları çalıştırıp bulguları iş etkisine göre önceliklendiren açık kaynak bir güvenlik platformudur.

Neden yükseliyor · 46 günde 492 yıldız ve son 28 günde 677 commit dikkat çekici bir ivmeye işaret ediyor; YC desteği ve browser, CLI, MCP birleşimi de görünürlüğü artırabilir. Ancak commitlerin tek farklı yazardan gelmesi, hareketliliğin geniş topluluk katkısı anlamına gelmediğini gösteriyor.

Autonomous AI pentesting agents — real-time reconnaissance, vulnerability detection, and exploitation orchestration. Go + TypeScript.

Deterministic Android unpacking pipeline for AI agents: zero-token mechanical analysis, LLM-outsourced bottlenecks, Rust-side verdicts.

ERA – Evolve & Reverse Engineer Anything: an AI-powered reverse-engineering agent built on REA, with a controlled, evidence-driven Improveme

Hardware hacker’s flying probe automation stack for agent-driven target discovery, microscope mapping, safety-monitored CNC motion, probe

Chinese-first LLM jailbreak benchmark & red-team toolkit | 中文LLM安全评测框架 | by POJIA.AI

Uncensored AI coding agent in your terminal. Does the work, skips the sermon. 256K context, crypto-only billing, no card on file.

Windsurf Cascade Bypass Methods 2026: Advanced Protobuf & MITM Exploit Repo

小liu破甲 - Codex CLI jailbreak prompt and one-click deploy tool. 基于CTF框架的Codex破甲提示词与一键部署工具

LLMs have ended reverse engineering as a high-barrier skill. Case study: reconstructing 6/7 custom HTTP signing parameters of Douyin (TikTok

An open database of real-world AI agent incidents.

Claude Code v2.1.88 source analysis: system prompts, 87 hidden feature flags, cost optimization, architecture diagrams, undercover mode, and

We read all 512K lines of Claude Code's accidentally exposed source. 82 docs, 15 diagrams, every subsystem mapped — from the hidden YOLO saf

High-assurance, artifact-agnostic prompt for authorized security labs, exploit-chain research, PoC validation, and evidence-led reporting.

Local-first offensive security engagement platform for mapping assets, credentials, sessions, pivots, findings, evidence, and attack paths.

A Model Context Protocol (MCP) server that gives AI agents live, stateful Windows RPC attack-surface research, built on James Forshaw's NtOb

OnionClaw™ — OpenClaw skill for full Tor dark web access: search .onion sites, fetch hidden services, rotate identity, run OSINT pipeline, c

Free open-source OSINT and investigation toolkit for Claude Code: case management, 54 OSINT enrichment adapters, structured analytic techniq

OSINT knowledge base + MCP server for autonomous AI agents — Claude Code, Cursor, Kimi K3, recon & threat intel playbooks.

OSINT tool for precise version fingerprinting of open-source web software

Extract the embedded web frontend (HTML/CSS/JS) out of a compiled Tauri app binary. The asar-extract Tauri never had.

High-fidelity Claude Fable 5 (Mythos) environment emulation and automated multi-agent jailbreak (Pack Hunt) research laboratory.

通用逆向分析 KernelSU 模块：手机侧只做原子能力，智能全在 PC 侧（Claude Code + MCP）。仅限授权安全研究与教育用途。

A modular browser telemetry & fingerprinting research toolkit — explores Web APIs, device metadata, sensors, geolocation, media, and storage

Outside-in replication of Anthropic's Mythos Preview / Project Glasswing — open-source agentic vulnerability-discovery scaffold on Claude Op

Suijin is an open-source fully autonomous red teaming / web app pentesting framework for AI models. Powered by LangGraph, designed for bug b

🛡 The reference playbook for pentesting AI chatbots & LLM-powered apps in one place. Ready-to-use payloads covering the full OWASP LLM Top 1

OpenAI Codex 自定义系统提示词注入工具 | System Prompt Injection | CTF Security Research Red Team | Auto Install | Cross-Platform

TRAE CN（原 TRAE SOLO CN）桌面客户端凭据存储逆向 — 4/4 解密，98 个 API，ECDSA P-256 Token 刷新，每日签到。2026-10：Trae CN 1.107.1 重测通过，OAuth 迁移 api.trae.com.cn，新增网页授权登

Run Claude Code CLI with DeepSeek API — cheap high-volume automation for security research & bug bounty

Searchable archive of Windows kernel symbols, struct layouts, function RVAs and byte patterns extracted from Microsoft PDBs and pinned to ex

SunnyDayBPF: eBPF-based post-syscall user-buffer telemetry deception research by Azizcan Daştan

AVE - Agentic Vulnerability Enumeration. The open behavioral classification standard for agentic AI components.

Mythos-class autonomous vulnerability discovery from any strong coding model (Opus 4.7, GPT-5.5, DeepSeek V4, Gemini). Long-running, email-d

AI-powered security research assistant for Claude Code — structured assessment workflows, tool orchestration, and professional reporting acr

Terminal-native passive OSINT CLI for local-first public-source investigations, evidence correlation, and case management.

Reproducible research on the persistence and recurrence of security misconfigurations in public Terraform repositories.

Runtime feature research for Black Desert x64: FOV unlock, render-scale supersampling (up to 400%), actor chams, and gathering ESP. Source,