MCP MeasureMCP Measure

索引已觀測

GoAI Moat Brand Intel Memory

Persistent business-intelligence memory for AI agents.

我該用這個嗎

品質與安全性

A
說明品質
93%
結構描述完整度
97%
命名品質
87%
汙染風險
100%
權限相符程度
100%
協定合規性
100%

根據工具定義與協定合規性的自動化分析。

上下文成本

~271Token（工具定義）
~870 B典型回應大小
極小的注意力影響（128k 上下文的 0.21%）

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": {
    "brand-intel-memory": {
      "url": "https://memory.mcp.goaimoat.com/mcp"
    }
  }
}

遠端端點

https://memory.mcp.goaimoat.com/mcpstreamable-http

它能做什麼

工具清單

工具（3）

🟢 唯讀🟡 寫入🔴 刪除⚪ 未知
⚪remember(brand_name, facts, category)

Store key facts about a brand for later recall.

輸入結構描述

{
  "type": "object",
  "properties": {
    "brand_name": {
      "type": "string",
      "description": "The brand/company name."
    },
    "facts": {
      "type": "string",
      "description": "Key facts discovered (free text, e.g. \"sells phone cases, 108 countries, AI-visible on iMore\")."
    },
    "category": {
      "default": "",
      "type": "string",
      "description": "Optional category (e.g. \"phone case\", \"DTC fashion\")."
    }
  },
  "required": [
    "brand_name",
    "facts"
  ],
  "additionalProperties": false
}

輸出結構描述

{
  "type": "object",
  "additionalProperties": true
}
🟢recall(brand_name)

Retrieve previously stored facts about a brand.

輸入結構描述

{
  "type": "object",
  "properties": {
    "brand_name": {
      "type": "string",
      "description": "The brand/company name to recall."
    }
  },
  "required": [
    "brand_name"
  ],
  "additionalProperties": false
}

輸出結構描述

{
  "type": "object",
  "additionalProperties": true
}
🟢list_brands(category)

List all remembered brands, optionally filtered by category.

輸入結構描述

{
  "type": "object",
  "properties": {
    "category": {
      "default": "",
      "type": "string",
      "description": "Optional category filter."
    }
  },
  "additionalProperties": false
}

輸出結構描述

{
  "type": "object",
  "additionalProperties": true
}

社群

為此伺服器評分

證據

近期觀測

已驗證未記錄版本3 個工具
已驗證未記錄版本3 個工具