From 7a07201b0c1879c149be2f6ff11d8520167bf7c0 Mon Sep 17 00:00:00 2001 From: "copilot-swe-agent[bot]" <198982749+Copilot@users.noreply.github.com> Date: Thu, 29 Jan 2026 09:32:34 +0000 Subject: [PATCH 1/2] Initial plan From 5bbdaef4df1f61bcba46764ce8e02f7dd0bb737e Mon Sep 17 00:00:00 2001 From: "copilot-swe-agent[bot]" <198982749+Copilot@users.noreply.github.com> Date: Thu, 29 Jan 2026 09:51:42 +0000 Subject: [PATCH 2/2] docs: add AI-Infra-Guard to BlackIce tools Co-authored-by: boy-hack <18695984+boy-hack@users.noreply.github.com> --- ubuntu/blackice/README.md | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/ubuntu/blackice/README.md b/ubuntu/blackice/README.md index 86d4512..df8fc44 100644 --- a/ubuntu/blackice/README.md +++ b/ubuntu/blackice/README.md @@ -17,7 +17,7 @@ For more details, see our paper [BlackIce: A Containerized Red Teaming Toolkit f ## Integrated Toolset -BlackIce integrates 14 widely-adopted open-source AI red teaming tools, chosen for their effectiveness and broad adoption across industry-leading AI security teams. Tools included cover a wide spectrum of evaluation capabilities, from basic static vulnerability assessments to highly customizable dynamic attack frameworks. +BlackIce integrates 15 widely-adopted open-source AI red teaming tools, chosen for their effectiveness and broad adoption across industry-leading AI security teams. Tools included cover a wide spectrum of evaluation capabilities, from basic static vulnerability assessments to highly customizable dynamic attack frameworks. | Tool | Organization | Stars | Type | Environment | Source | License | |-------------------------------------------|----------------|-------|---------|-------------|--------|-----------| @@ -28,6 +28,7 @@ BlackIce integrates 14 widely-adopted open-source AI red teaming tools, chosen f | [ART](#art) | IBM | 5.6K | Dynamic | Isolated | PyPI | MIT | | [Giskard](#giskard) | Giskard | 4.9K | Hybrid | Isolated | PyPI | Apache 2.0| | [CyberSecEval](#cyberseceval) | Meta | 3.8K | Static | Isolated | GitHub | MIT | +| [AI-Infra-Guard](#ai-infra-guard) | Tencent | 2.9K | Hybrid | Isolated | GitHub | MIT | | [PyRIT](#pyrit) | Microsoft | 2.9K | Dynamic | Global | PyPI | MIT | | [EasyEdit](#easyedit) | ZJUNLP | 2.6K | Dynamic | Isolated | GitHub | MIT | | [Promptmap](#promptmap) | - | 1K | Static | Isolated | GitHub | GPL-3.0 | @@ -287,6 +288,20 @@ Make sure to replace: Check out `cyberseceval --help`, the [official documentation](https://meta-llama.github.io/PurpleLlama/), or the [GitHub Repository](https://github.com/meta-llama/PurpleLlama/tree/main/CybersecurityBenchmarks) for more information regarding the different options available. +## AI-Infra-Guard +AI-Infra-Guard is an AI red teaming platform from Tencent Zhuque Lab that integrates AI infra vulnerability scanning, MCP Server risk scanning, and jailbreak evaluation to help teams assess AI security risks. + +**Quick start (Docker)** +```bash +git clone https://github.com/Tencent/AI-Infra-Guard.git +cd AI-Infra-Guard +docker-compose -f docker-compose.images.yml up -d +``` + +Once running, open `http://localhost:8088` to access the web interface. + +Refer to the [official documentation](https://tencent.github.io/AI-Infra-Guard/) or the [GitHub Repository](https://github.com/Tencent/AI-Infra-Guard) for more details. + ## PyRIT PyRIT is an open-source framework designed for evaluating and red teaming generative AI systems. It allows users to configure various components—such as Prompt Targets, Orchestrators, Scorers, and Converters—to structure, execute, and analyze interactions with AI models.