CRM CSV Preflight (Sales List Rescue)

Detect-only CSV check before CRM import: duplicate, invalid and role emails, blanks, phones, dates.

我該用這個嗎

品質與安全性

B
說明品質
80%
結構描述完整度
65%
命名品質
80%
汙染風險
60%
權限相符程度
100%
協定合規性
100%

發現項目(5)

  • HIGHTool poisoning patterns detected
  • MEDIUMTool 'preflight_csv' description contains placeholder text在 preflight_csv 中
  • MEDIUMTool 'preflight_sample' description contains placeholder text在 preflight_sample 中
  • INFOTool description contains placeholder or incomplete text在 preflight_csv 中
  • INFOTool description contains placeholder or incomplete text在 preflight_sample 中

根據工具定義與協定合規性的自動化分析。

上下文成本

~569Token(工具定義)
~1.4 KB典型回應大小
極小的注意力影響(128k 上下文的 0.44%)

這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高,可用於其他工作的注意力就越少。

安裝

一鍵安裝

將以下內容加入你的 `claude_desktop_config.json` 檔案:

{
  "mcpServers": {
    "crm-csv-preflight": {
      "url": "https://pureservices.store/mcp"
    }
  }
}

遠端端點

https://pureservices.store/mcpstreamable-http

它能做什麼

工具清單

工具(2)

🟢 唯讀🟡 寫入🔴 刪除⚪ 未知
🟢preflight_csv(csv, default_country_code, email_column, phone_column, required_columns)

Checks a contact or lead CSV (passed as text, up to 500 data rows and 64 KiB) before a CRM import and reports what it finds: duplicate emails (matched ignoring case and surrounding spaces), invalid email syntax, role or shared inboxes (info@, sales@ and similar), blank required fields (placeholders like N/A, null and - count as blank), phone numbers not written as + country code and digits only (with a suggested form when the fix is mechanical), country and state spelling drift, mixed or ambiguous date formats, and file problems (garbled characters, ragged rows, formula-like cells, stray spaces, header issues). Detection only: it changes nothing, returns no cleaned file, and does not store or log your CSV. Pattern-based; it does not check that mailboxes exist. Same checks as the free Sales List Rescue CSV preflight checker on pureservices.store.

輸入結構描述

{
  "type": "object",
  "properties": {
    "csv": {
      "type": "string",
      "description": "The full CSV text including the header row. Comma, semicolon, tab or pipe delimited (auto-detected)."
    },
    "default_country_code": {
      "type": "string",
      "enum": [
        "1",
        "44",
        "61",
        "49",
        "33",
        "91",
        ""
      ],
      "default": "1",
      "description": "Calling code assumed for phone numbers without a + prefix: 1 (US/Canada), 44, 61, 49, 33, 91, or \"\" for no default."
    },
    "email_column": {
      "type": "string",
      "description": "Header name of the email column. Optional; auto-detected when omitted."
    },
    "phone_column": {
      "type": "string",
      "description": "Header name of the phone column. Optional; auto-detected when omitted."
    },
    "required_columns": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Header names that must not be blank. Optional; defaults to the detected email, first name, last name and company columns."
    }
  },
  "required": [
    "csv"
  ],
  "additionalProperties": false
}
🟢preflight_sample

Runs the same checks as preflight_csv on the checker's built-in sample CSV (fake placeholder data) and returns the sample plus its report, so you can see the report format without sending a file. Detection only; does not store or log your CSV.

輸入結構描述

{
  "type": "object",
  "properties": {},
  "additionalProperties": false
}

社群

為此伺服器評分

證據

近期觀測

已驗證未記錄版本2 個工具