LeakData

Check breach exposure for your verified email and check locally computed password hash prefixes.

我该使用它吗

质量与安全性

A
描述质量
100%
模式完整度
95%
命名质量
50%
投毒风险
100%
权限匹配度
100%
协议合规性
100%

发现(2)

  • LOWTool 'leakdata.search' doesn't follow camelCase/snake_case在 leakdata.search 中
  • LOWTool 'leakdata.password_prefix_check' doesn't follow camelCase/snake_case在 leakdata.password_prefix_check 中

基于对工具定义和协议合规性的自动分析。

上下文开销

~802token 数(工具定义)
~2.8 KB典型响应大小
对注意力有中等影响(占 128k 上下文窗口的 0.63%)

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": {
    "leakdata-mcp": {
      "url": "https://leakdata.io/mcp"
    }
  }
}

远程端点

https://leakdata.io/mcpstreamable-http

它能做什么

工具清单

工具(2)

🟢 只读🟡 写入🔴 删除⚪ 未知
🟢leakdata.search(type, query)

Use this when the linked LeakData user asks to check their verified primary email. For a current check, call this tool and report only its returned results. Never present earlier conversation results as a new check; if the call fails, say the check could not be completed. For their own email without an address, set type to email and omit query: the server resolves the verified primary email from the linked account. If the user supplies an email address, preserve it in query for ownership validation. Never guess an email address or recover it from earlier conversations when query can be omitted. Domain checks are not available through this app; explain this without calling a tool. Arbitrary third-party identifiers, usernames, phone numbers, raw leak records, and credentials are not returned.

输入模式

{
  "type": "object",
  "properties": {
    "type": {
      "type": "string",
      "enum": [
        "email"
      ]
    },
    "query": {
      "type": "string",
      "minLength": 1,
      "maxLength": 320,
      "description": "Omit for the linked account's verified primary email. If an email is supplied, it must match that verified primary email."
    }
  },
  "required": [
    "type"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "found": {
      "type": "boolean"
    },
    "breachCount": {
      "type": "integer",
      "minimum": 0
    },
    "type": {
      "type": "string",
      "enum": [
        "email"
      ]
    },
    "breaches": {
      "type": "array",
      "maxItems": 20,
      "items": {
        "type": "object",
        "additionalProperties": false,
        "required": [
          "source",
          "breachDate",
          "dataTypes"
        ],
        "properties": {
          "source": {
            "type": "string",
            "minLength": 1,
            "maxLength": 200
          },
          "breachDate": {
            "anyOf": [
              {
                "type": "string",
                "format": "date"
              },
              {
                "type": "null"
              }
            ]
          },
          "dataTypes": {
            "type": "array",
            "maxItems": 30,
            "items": {
              "type": "string",
              "minLength": 1,
              "maxLength": 100
            }
          },
          "affectedUsers": {
            "type": "integer",
            "minimum": 0
          }
        }
      }
    },
    "truncated": {
      "type": "boolean"
    }
  },
  "required": [
    "found",
    "breachCount",
    "type",
    "breaches",
    "truncated"
  ],
  "additionalProperties": false
}
🟢leakdata.password_prefix_check(prefix)

Call this only when the current user message supplies a locally computed five-character SHA-1 hash prefix and asks to check it. Never invent a prefix or reuse one from an earlier message. For general questions about password checks, explain local SHA-1 computation and sharing only its first five hexadecimal characters without calling a tool. Never accept a plaintext password or complete hash. Report only the aggregate values actually returned by this call. Results describe a prefix group only. A matching prefix does not establish whether a particular password is exposed. Do not report confirmed password exposure or recommend changing a password based only on this result. An empty group does not prove that a password is safe.

输入模式

{
  "type": "object",
  "properties": {
    "prefix": {
      "type": "string",
      "pattern": "^[0-9A-Fa-f]{5}$",
      "minLength": 5,
      "maxLength": 5,
      "description": "Five-character SHA-1 hash prefix supplied in the current user message. Never infer it from earlier messages."
    }
  },
  "required": [
    "prefix"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "found": {
      "type": "boolean",
      "description": "Whether any known hashes share this five-character prefix. A match does not establish that a particular password is exposed."
    },
    "candidateHashCount": {
      "type": "integer",
      "minimum": 0,
      "description": "Number of distinct known hashes sharing the prefix, not matches for one specific password."
    },
    "totalOccurrences": {
      "type": "integer",
      "minimum": 0,
      "description": "Total occurrences across all hashes in the prefix group, not an exposure count for one specific password."
    }
  },
  "required": [
    "found",
    "candidateHashCount",
    "totalOccurrences"
  ],
  "additionalProperties": false
}

社区

评价此服务器

证据

最近观测

已验证未记录版本2 个工具
已验证未记录版本2 个工具
已验证未记录版本2 个工具