A full-stack AI red teaming platform for self-examining security risks across AI infrastructure, agents, and MCP servers. It can be self-hosted locally via Docker Compose or Python, but certain scanning and jailbreak evaluation features require a remote LLM API key.
Project overview
Provides a web interface and CLI for scanning AI services and Agent Skills for vulnerabilities and jailbreaks.
Project type
MCP
Deployment
Refer to project documentation
License
Apache-2.0
Best for
Developers, operations teams, and AI engineers performing security risk self-examination on internal AI services, provided deployment is restricted to private networks due to the absence of an authentication mechanism.
Key capabilities
A standalone CLI tool that scans local Skill project directories and can be integrated into CI/CD pipelines.
User-friendly UI with one-click scanning and real-time progress tracking.
Limitations and risks
Currently lacks an authentication mechanism and should not be deployed on public networks.
Deploying on public networks carries risk due to the lack of authentication.
Getting started
Navigate to the project directory, run docker-compose using the pre-built images file, and access the web interface at localhost:8088.
Evidence and sources
GitHub project description: 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.
README: 🚀 AI Red Teaming Platform by Tencent Zhuque Lab
README: **[ClawScan(OpenClaw Security Scan)](https://matrix.tencent.com/clawscan)** | Supports one-click evaluation of OpenClaw security risks. It detects insecure configuration…