MCP Server Security Review

高级security最低上下文：32K

Reviews Model Context Protocol servers for excessive permissions, unsafe tool design, prompt injection exposure, secret handling, authorization gaps, and auditability.

mcpsecurityprompt-injectionai-agents

使用场景

  • MCP security assessments
  • Tool permission reviews
  • Agent hardening

示例提示词

Perform a security review of this MCP server. Find excessive privileges, injection paths, unsafe side effects, secret exposure, and missing authorization controls; propose mitigations.

推荐模型

claude-opus-4-6gpt-5gemini-2-5-pro

兼容工具

claude-codecursorkiroany

模态

输入: text, code, file
→
输出: text, code

相关 Skills

security-auditprompt-injection-tester

作者

OpenModels Community

@openmodelsrun