🛡️ A.I.G（AI-Infra-Guard）

Name: 🛡️ A.I.G（AI-Infra-Guard）
Author: Tencent

@Tencent

🛡️ A.I.G（AI-Infra-Guard）について

A full-stack AI Red Teaming platform securing AI ecosystems via OpenClaw Security Scan, Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM jailbreak evaluation.

基本情報

カテゴリ

AI とエージェント

ライセンス

Apache-2.0

ランタイム

python

トランスポート

stdio

公開者

Tencent

設定

標準の設定はありません

このサーバーの README には解析可能な MCP 設定ブロックが含まれていません。インストール手順はリポジトリをご確認ください。

リポジトリ

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is A.I.G (AI-Infra-Guard)?

A.I.G (AI-Infra-Guard) is an AI Red Teaming Platform developed by Tencent Zhuque Lab. It integrates capabilities such as ClawScan, Agent Scan, AI infrastructure vulnerability scanning, MCP Server & Agent Skills scanning, and Jailbreak Evaluation. The platform is designed for developers and security teams to perform comprehensive self-examination of AI security risks across models, agents, and infrastructure.

How to use A.I.G (AI-Infra-Guard)?

Deploy with Docker: clone the repository, run docker-compose -f docker-compose.images.yml up -d, then access the web interface at http://localhost:8088. For a one‑click installation, use curl https://raw.githubusercontent.com/Tencent/AI-Infra-Guard/refs/heads/main/docker.sh | bash. A.I.G can also be invoked from OpenClaw via the aig-scanner skill by installing it with clawhub install aig-scanner and configuring AIG_BASE_URL. The platform currently has no authentication and must not be deployed on public networks.

Key features of A.I.G (AI-Infra-Guard)

ClawScan: One‑click evaluation of OpenClaw security risks.
Agent Scan: Multi‑agent framework for AI agent workflow security.
MCP Server & Agent Skills scan: Detects 14 major security risk categories.
AI infra vulnerability scan: Identifies 100+ AI components and 1900+ CVEs.
Jailbreak Evaluation: Assesses prompt security with multiple attack methods.
Free and open source under Apache 2.0 license.

Use cases of A.I.G (AI-Infra-Guard)

Scan a locally running vLLM instance for

「AI とエージェント」の他のコンテンツ

Model Context Protocol for Unreal Engine

chongdashu

Enable AI assistant clients like Cursor, Windsurf and Claude Desktop to control Unreal Engine through natural language using the Model Context Protocol (MCP).

2.0K

1Panel

1Panel-dev

🔥 1Panel is a modern, open-source VPS control panel — and the only one with native AI agent support. Run Ollama models, deploy OpenClaw agents, and manage your entire server stack from one clean web interface.

36.1K

Solon Ai

opensolon

Java AI application development framework (supports LLM-tool,skill; RAG; MCP; Agent-ReAct,Team-Agent). Compatible with java8 ~ java25. It can also be embedded in SpringBoot, jFinal, Vert.x, Quarkus, and other frameworks.

416