AISecurityLab/hackagent

HackAgent is an open-source security toolkit to detect vulnerabilities of your AI Agents

What it solves

HackAgent is a security red-teaming toolkit designed to find vulnerabilities in AI agents. It addresses the risk of prompt injection, jailbreaking, goal hijacking, and tool misuse, which traditional security tools cannot detect.

How it works

It uses a modular pipeline consisting of an Attack Engine that orchestrates various research-backed attack techniques (such as PAIR, TAP, and AutoDAN-Turbo), a Generator LLM to create adversarial prompts, and a Judge LLM to evaluate if the safety measures were bypassed. This process is automated to test target agents built on frameworks like LangChain, OpenAI SDK, LiteLLM, and Google ADK.

Who it’s for

Security researchers, developers, and AI safety practitioners who need to evaluate and strengthen the robustness of their AI agents.

Highlights

  • Automated Red-Teaming: Uses a variety of advanced attack techniques to automate the discovery of security flaws.
  • Broad Framework Support: Compatible with major AI agent frameworks including LangChain and OpenAI SDK.
  • Flexible Reporting: Offers both a local SQLite-based dashboard and a cloud-based reporting platform.
  • Easy Deployment: Available as a Python SDK, CLI, or standalone binaries for multiple platforms.

Related

  • Project
    Kritt-ai/open-kritt

    An open-source security research platform that orchestrates AI agents to find, validate, and rank vulnerabilities in codebases using focused, parallelized workflows.

  • Project
    Tencent/AI-Infra-Guard

    An AI red teaming platform that provides automated security scanning for AI infrastructure, agent skills, and LLM jailbreak robustness.

  • Project
    asamassekou10/ship-safe

    An AI security scanner that detects vulnerabilities in application code, AI agents, and supply chains, specifically targeting risks like prompt injection and agent hijacking.

  • Project
    confident-ai/deepteam

    An open-source red teaming framework for LLM systems that simulates adversarial attacks to uncover vulnerabilities and provides guardrails for production security.

  • Project
    tophant-ai/aibeat

    A security evaluation framework for generative AI that provides evidence-based safety testing for LLMs, RAG apps, and AI agents.