Yeti-791/Tsec-Hackathon

腾讯云智能渗透黑客松 Official repository of Tencent Cloud Intelligent Penetration Hackathon. Showcasing top open-source projects of LLM-based autonomous penetration agents, including multi-agent collaboration, automated penetration, AI-driven offensive security, and intelligent attack-defense solutions.

What it solves

Building an AI-powered penetration testing agent is hard: teams reinvent architecture, tooling, and evaluation strategies. This repository acts as a central archive for the Tencent Cloud Intelligent Penetration Hackathon (TCH), collecting defense slides, event materials, and links to the open-source projects of the top 20 teams from two editions. It gives newcomers and practitioners a single starting point to learn how successful LLM-driven penetration agents are designed, implemented, and evaluated.

How it works

The repository is organized as a navigation index. It lists the top-ranking teams from the first and second hackathons, each with a short highlight (e.g., "Manager+Solver+Observer three-layer decoupling" or "blackboard system + ant colony algorithm"), a link to their public repository, and in many cases a link to their recorded defense video and presentation PDF. The README also points to official event pages, a public model-gateway log report, a set of post-event analysis articles, and a separate curated collection of papers, benchmarks, and commercial products in the AI-offensive-security space.

Who it's for

  • Cybersecurity researchers and engineers who want to study real-world intelligent penetration agents
  • Students and teams preparing for AI security competitions
  • Anyone looking for a structured overview of the current open-source landscape in LLM-driven offensive security

Highlights

  • Aggregates the top 20 open-source projects from each of two hackathon editions, covering multi-agent collaboration, autonomous planning, and scenario-aware black-box penetration
  • Includes official PPTs, videos, and public chat logs from the competition platform for deep analysis
  • Features high-star repositories like CyberStrikeAI (6.6k stars), Cairn (2.8k stars), and LuaN1aoAgent (1.3k stars)
  • Links to a comprehensive learning-assets document that surveys AI penetration testing, LLM red teaming, benchmarks, and commercial products
  • All materials are for educational use and clearly marked with a disclaimer against illegal applications

Related

  • Project
    kody-hq/kody

    Kody is a Cloudflare Workers‑based platform that provides persistent memory, credential storage, code packages, and job scheduling for AI agents via the Model Context Protocol (MCP). It offers a Remix UI, OAuth‑protected endpoints, and a four‑worker architecture (origin, platform, runtime, jobs) to serve individual or team accounts. The repo is a full‑stack TypeScript monorepo (Nx) with local dev support, extensive docs, and a functional‑source license.

  • Project
    RUC-NLPIR/Awesome-Long-Horizon-Agents

    A GitHub repository that curates 750+ papers, benchmarks, and resources about long-horizon AI agents, organized by themes like evolution, harness engineering, optimization, and applications. It accompanies a survey paper and serves as a map of the field.

  • Project
    AgriciDaniel/chatgpt-seo

    SEO-pro is a ChatGPT plugin that turns page URLs, drafts, or Search Console exports into sourced, prioritized SEO fixes with verification steps. It works directly inside ChatGPT, requires no API key or external SEO account, and uses only the evidence you provide.

  • Project
    langchain-ai/react-agent

    A LangGraph template for implementing a ReAct agent that iteratively reasons and executes tool-based actions to solve complex user queries.