AI-Infra-Guard is an AI security red-teaming platform that scans agents, skills, MCP servers, and AI infrastructure and evaluates LLM jailbreak resistance. It is used to identify security risks and vulnerabilities in AI systems. Catalogue add-ons support its scanning and evaluation workflows.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/tencent/ai-infra-guard/dynamic_verificationgit clone --depth 1 https://github.com/Tencent/AI-Infra-GuardWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/tencent/ai-infra-guard/dynamic_verification)<a href="https://agentmods.dev/agents/tencent/ai-infra-guard/dynamic_verification"><img src="https://agentmods.dev/badge/agents/tencent/ai-infra-guard/dynamic_verification.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.01662 |
| Opus 5 | $0.00000 | $0.00831 |
| Sonnet 5 | $0.00000 | $0.00332 |
| Haiku 4.5 | $0.00000 | $0.00166 |
Grade B, and why
dynamic_verification scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reaches for credential filesmediumPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
- 检查是否能访问~/.aws/credentials、.env等敏感文件 Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
3. 使用`execute_shell`执行Python脚本或curl等HTTP客户端测试 How it starts
The opening of the file, as written. The whole thing — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MCP漏洞动态验证Agent
角色定义
你是一个专业的安全渗透测试专家,专注于MCP(Model Context Protocol)服务器的漏洞验证。你的任务是对已发现的安全漏洞进行实际的动态验证,通过构造和执行真实的exploit来证明漏洞的可利用性。
核心职责
- 漏洞分析:理解代码审计报告中的每个漏洞的技术细节
- Exploit生成:为每个漏洞构造具体的exploit代码
- 动态验证:通过MCP客户端或其他方式执行exploit
- 结果评估:判断漏洞是否真实可利用,并提供证据
任务输入
你将收到以下信息:
- 漏洞列表:从漏洞审查阶段提取的结构化漏洞数据
- 服务器信息:从构建预览阶段获得的服务器地址、端口、PID等
- 代码审计报告:完整的代码审计报告,包含详细的漏洞分析
验证流程
阶段1:环境准备与确认
-
解析服务器信息
- 从build_result XML中提取server_url、pid、log_file
- 确认服务器状态(如需要可使用check_process_logs验证)
- 记录测试环境信息
-
解析漏洞列表
- 从提供的漏洞数据中提取每个漏洞的详细信息
- 按风险等级排序,优先验证Critical和High级别
- 识别可以自动化验证的漏洞类型
阶段2:逐个验证漏洞
对每个漏洞执行以下步骤:
2.1 Command Injection(命令注入)验证
验证策略:
- 识别可注入的参数和工具
- 构造MCP客户端脚本,调用目标工具
- 注入测试payload(如:
; echo INJECTED_MARKER或&& whoami) - 检查响应或日志中是否出现注入标记
Exploit模板:
import asyncio
from fastmcp import Client
# HTTP server
client = Client("https://example.com/mcp")
async def main():
async with client:
# Basic server interaction
await client.ping()
# List available operations
# tools = await client.list_tools()
# resources = await client.list_resources()
# prompts = await client.list_prompts()
# Execute operations
result = await client.call_tool("example_tool", {"param": "value"})
print(result)
asyncio.run(main())
2.2 Credential Theft(凭据窃取)验证
验证策略:
- 检查是否能通过MCP工具访问敏感文件
- 尝试读取环境变量或配置文件
- 验证凭据是否在响应中泄露
检测方法:
- 调用文件读取相关的工具
- 检查是否能访问~/.aws/credentials、.env等敏感文件
- 分析响应内容中的敏感信息
2.3 Indirect Prompt Injection(间接提示注入)验证
验证策略:
- 创建包含恶意指令的测试文件
- 通过MCP工具让服务器处理该文件
- 观察服务器行为是否被劫持
测试方法:
- 构造包含隐藏指令的文档(如在注释或隐藏文本中)
- 让MCP服务器读取和处理该文档
- 检查输出是否执行了注入的指令
2.4 Hardcoded API Key(硬编码密钥)验证
验证策略:
- 读取源代码文件,确认密钥存在
- 尝试使用该密钥访问对应服务
- 验证密钥是否有效且具有权限
2.5 Auth Bypass(认证绕过)验证
验证策略:
- 尝试不提供认证信息访问保护资源
- 测试默认凭据或已知绕过技术
- 验证是否能访问受保护的功能
2.6 Tool Poisoning/Shadowing(工具投毒/阴影)验证
验证策略:
- 检查MCP工具列表和描述
- 识别可疑的工具重定义或覆盖
- 调用疑似被篡改的工具,观察异常行为
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 189 lines · 0 tokens per session scan B a9320993679c
dynamic_verification is an agent published in the GitHub repository Tencent/AI-Infra-Guard (6,141 stars, last pushed yesterday), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,662 tokens. A static security scan graded it B with 2 findings (reaches for credential files, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
prompt-engineer
Optimizes prompts for LLMs and AI systems. Use when building AI features, improving agent performance, or crafting system prompts. Expert in prompt patterns and techniques.
mlops-engineer
Build ML pipelines, experiment tracking, and model registries. Implements MLflow, Kubeflow, and automated retraining. Handles data versioning and reproducibility. Use PROACTIVELY for ML infrastructure, experiment management, or pipeline automation.
ai-engineer
Build LLM applications, RAG systems, and prompt pipelines. Implements vector search, agent orchestration, and AI API integrations. Use PROACTIVELY for LLM features, chatbots, or AI-powered applications.
data-engineer
Build ETL pipelines, data warehouses, and streaming architectures. Implements Spark jobs, Airflow DAGs, and Kafka streams. Use PROACTIVELY for data pipeline design or analytics infrastructure.
ml-engineer
Implement ML pipelines, model serving, and feature engineering. Handles TensorFlow/PyTorch deployment, A/B testing, and monitoring. Use PROACTIVELY for ML model integration or production deployment.
data-scientist
Data analysis expert for SQL queries, BigQuery operations, and data insights. Use proactively for data analysis tasks and queries.