HomeProjectsArticlesPK BoardAbout
Explore Projects

AI-Infra-Guard

Active
GitHub Python Apache-2.0

Description

Tencent's full-stack AI red teaming platform integrating OpenClaw security scanning, agent scanning, skills scanning, MCP scanning, AI infrastructure scanning, and LLM jailbreak evaluation.

Key Features

  • ClawScan: OpenClaw security scanning for detecting vulnerabilities in OpenClaw configurations and skills
  • Agent Scan: vulnerability scanning across AI agents with WebSocket provider support
  • AI infrastructure vulnerability scan covering 68+ AI components with 600+ CVE rules
  • MCP Server and Agent Skills scan for supply chain security and configuration exposure detection
  • LLM jailbreak evaluation with 26 attack operators including single-turn and multi-turn attacks

Use Cases

💡 Self-assessment of AI deployment security posture across infrastructure, agents, and MCP services
💡 Supply chain security scanning to detect malicious packages and exposed configuration files
💡 LLM safety evaluation through jailbreak testing and prompt injection attack simulation
💡 Continuous security monitoring of AI components with Docker-based automated scanning pipelines

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (6.1k stars)
  • • Permissive open-source license (Apache-2.0)
  • • Responsive to issues, low backlog

Tags

ai-security red-teaming llm-security vulnerability-scanner safety-evaluation

Categories

📊 Observability 🛡️ Security & Guardrails

Quick Start

git clone https://github.com/Tencent/AI-Infra-Guard.git && cd AI-Infra-Guard && docker-compose -f docker-compose.images.yml up -d. Access the web UI at http://localhost:8088. From OpenClaw: clawhub install aig-scanner, then configure AIG_BASE_URL to your running service.
Visit GitHub Visit Website

Related Projects

Garak

9.1k · Python
Active A

NVIDIA's open-source LLM vulnerability scanner that automatically detects security issues in language models including safety vulnerabilities, hallucination tendencies, jailbreak risks, and prompt injection attacks.

llm-securityvulnerability-scannerllm-evaluation +2
  • · Automated LLM vulnerability scanning for hallucination, data leakage, prompt injection, and jailbreak detection
  • · Static, dynamic, and adaptive probe combinations for comprehensive security assessment
  • · Broad LLM support: Hugging Face, OpenAI, AWS Bedrock, Replicate, llama.cpp, and REST-accessible models
View full details → View on GitHub

Agentic Security

2.0k · Python
Active A

An open-source LLM vulnerability scanner and AI red teaming kit for automated security fuzzing of LLM applications, detecting jailbreaks, prompt injection, and adversarial attacks.

llm-securityred-teamingllm-fuzzer +2
  • · Multimodal attack probing across text, image, and audio inputs to test LLM robustness
  • · Multi-step jailbreak simulation with iterative attack sequences to uncover safety weaknesses
  • · Comprehensive fuzzing engine with randomized inputs to stress-test LLMs for edge cases
View on GitHub

OpenAI Evals

19.4k · Python
Stale B

OpenAI's framework for evaluating LLMs and LLM systems, providing an open-source registry of benchmarks and tools for systematic model assessment.

llm-evaluationbenchmarkevals +2
  • · Open-source registry of evals for testing different dimensions of LLM performance
  • · Custom eval creation using basic and model-graded templates without writing code
  • · Private evals support for evaluating LLM patterns in your workflow without exposing data
View full details → View on GitHub

Promptfoo

24.8k · TypeScript
Active A

Test and evaluate LLM prompts, agents, and RAG pipelines. Built-in red teaming and security evaluation for reliable AI applications.

testingevaluationred-teaming +2
  • · Automated LLM evaluations — Batch test prompts, models, and RAG pipeline output quality
  • · Red team security testing — Built-in vulnerability scanning and adversarial testing for LLM security
  • · Multi-model comparison — Side-by-side comparison of OpenAI, Anthropic, Azure, Bedrock, Ollama models
View full details → View on GitHub