AI-Infra-Guard

Red-team platform for agent, MCP, and skill infrastructure

AI Agents / Security · Apache-2.0 · 6.9k

Reviewed 2026-09-23 · last push 2026-10-19 · repo checked 2026-10-19

What's Good

Apache-2.0. The Tencent Zhuque Lab project bundles a self-hostable platform and CLI for probing AI infrastructure rather than treating prompts, MCP servers, and skills as trusted black boxes. Docker deployment keeps the test environment reproducible, while the breadth is useful for finding control-plane and tool-boundary failures before they reach an agent fleet.

The Catch

Vs Promptfoo and DeepEval: AI-Infra-Guard is an AI-infrastructure red-team platform, not only an eval runner; vs generic SAST, it understands agent/MCP/skill attack surfaces. Preserve the mandatory attribution — “Based on Tencent Zhuque Lab AI-Infra-Guard” — with the repository link in your deployment and derivative notices. Some tools also require an LLM_API_KEY, so a self-hosted scanner can still carry provider cost and credential risk.

Verdict

Apache-2.0 red-team platform for agent infrastructure. Worth running in a disposable, keyed environment; ship the attribution and scope every model credential.

Embed

Maintainer? Add the badge to your README. It is free, links to this review, and there is no paid version.

Markdown

[![Reviewed on Stack Gems](https://stackgems.com/badge/ai-infra-guard.svg)](https://stackgems.com/gems/ai-infra-guard)

HTML

<a href="https://stackgems.com/gems/ai-infra-guard"><img src="https://stackgems.com/badge/ai-infra-guard.svg" alt="AI-Infra-Guard reviewed on Stack Gems" height="20"></a>

AI Agents →

Related