ai-guardrails

Cloudflare Workers MCP server: ai-guardrails

我该使用它吗

质量与安全性

B
描述质量
76%
模式完整度
84%
命名质量
88%
投毒风险
80%
权限匹配度
100%
协议合规性
100%

发现(5)

  • HIGHTool poisoning patterns detected
  • MEDIUMTool 'redact_pii' description contains placeholder text在 redact_pii 中
  • LOWTool 'redact_pii' description lacks action verb在 redact_pii 中
  • LOWTool 'check_hallucination' description lacks action verb在 check_hallucination 中
  • INFOTool description contains placeholder or incomplete text在 redact_pii 中

基于对工具定义和协议合规性的自动分析。

上下文开销

~380token 数(工具定义)
~363 B典型响应大小
对注意力的影响极小(占 128k 上下文窗口的 0.30%)

这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高,可用于其他任务的注意力就越少。

安装

一键安装

将以下内容添加到你的 `claude_desktop_config.json` 文件中:

{
  "mcpServers": {
    "ai-guardrails": {
      "url": "https://api.lazy-mac.com/ai-guardrails/mcp"
    }
  }
}

远程端点

https://api.lazy-mac.com/ai-guardrails/mcpstreamable-http

它能做什么

工具清单

工具(5)

🟢 只读🟡 写入🔴 删除⚪ 未知
⚪detect_pii(text)

Detect personally identifiable information (email, phone, SSN, credit card, IP, URL, API key) in a text string.

输入模式

{
  "type": "object",
  "properties": {
    "text": {
      "type": "string",
      "description": "Text to scan for PII"
    }
  },
  "required": [
    "text"
  ]
}
⚪redact_pii(text, mode)

Redact PII from text. mode=mask replaces with [REDACTED]; mode=tag uses typed placeholders like [EMAIL].

输入模式

{
  "type": "object",
  "properties": {
    "text": {
      "type": "string"
    },
    "mode": {
      "type": "string",
      "enum": [
        "mask",
        "tag"
      ],
      "default": "mask"
    }
  },
  "required": [
    "text"
  ]
}
🟢check_safety(text)

Check text for harmful content using weighted keyword detection. Returns a 0–1 score and verdict (safe/caution/unsafe).

输入模式

{
  "type": "object",
  "properties": {
    "text": {
      "type": "string"
    }
  },
  "required": [
    "text"
  ]
}
🟢check_hallucination(text)

Heuristic scoring of LLM hedge phrases that correlate with hallucinated outputs. Returns score 0–1 and risk level.

输入模式

{
  "type": "object",
  "properties": {
    "text": {
      "type": "string"
    }
  },
  "required": [
    "text"
  ]
}
🟢full_check(text, checks)

Run pii, safety, and/or hallucination checks in one call. Default: pii + safety. Specify checks=["pii","safety","hallucination"] to run all.

输入模式

{
  "type": "object",
  "properties": {
    "text": {
      "type": "string"
    },
    "checks": {
      "type": "array",
      "items": {
        "type": "string",
        "enum": [
          "pii",
          "safety",
          "hallucination"
        ]
      },
      "default": [
        "pii",
        "safety"
      ]
    }
  },
  "required": [
    "text"
  ]
}

社区

评价此服务器

证据

最近观测

已验证未记录版本5 个工具
已验证未记录版本5 个工具
已验证未记录版本5 个工具