Kayzen Analytics APIと連携するためのモデルコンテキストプロトコル（MCP）サーバー実装。このパッケージにより、AIモデルは標準化されたインターフェースを介してKayzen広告キャンペーンデータにアクセスし、分析できるようになります。
自動認証: 自動更新メカニズムを備えた組み込みトークン管理
レポート管理: Kayzen分析レポートへの簡単なアクセス
エラー処理: APIインタラクションの包括的なエラー処理
TypeScript サポート: 型定義を備えた完全な TypeScript 実装
環境ベースの構成:環境変数を使用した簡単なセットアップ
Related MCP server: Amazon Ads API MCP SDK
npm install @feedmob-ai/kayzen-mcpKayzen の資格情報を使用して.envファイルを作成します。
KAYZEN_USERNAME=your_username
KAYZEN_PASSWORD=your_password
KAYZEN_BASIC_AUTH=your_basic_auth_token
KAYZEN_BASE_URL=https://api.kayzen.io/v1 # Optional, defaults to this valueimport { KayzenMCPServer } from '@feedmob-ai/kayzen-mcp';
const server = new KayzenMCPServer();
server.start();list_reportsKayzen Analytics から利用可能なすべてのレポートを一覧表示します。
入力: なし
戻り値: 次のものを含むレポート オブジェクトの配列:
id : レポート識別子
name : レポート名
type : レポートタイプ
const reports = await server.tools.list_reports();get_report_results特定のレポートの結果を取得します。
入力:
report_id (文字列、必須): 取得するレポートのID
start_date (文字列、オプション): 開始日（YYYY-MM-DD形式）
end_date (文字列、オプション): 終了日（YYYY-MM-DD形式）
戻り値: レポートデータとメタデータ
const results = await server.tools.get_report_results({
report_id: 'report_id',
start_date: '2024-01-01', // optional
end_date: '2024-01-31' // optional
});analyze_report_results (プロンプト)レポート結果を分析し、洞察を提供します。
入力:
report_id (文字列): 分析するレポートのID
分析には以下が含まれます:
パフォーマンス指標
主な傾向
最適化の領域
異常なパターンや異常
Claude Desktop でこれを使用するには、 claude_desktop_config.jsonに以下を追加します。
{
"mcpServers": {
"github": {
"command": "npx",
"args": [
"-y",
"@feedmob-ai/kayzen-mcp"
],
"env": {
"KAYZEN_USERNAME": "username",
"KAYZEN_PASSWORD": "pasword",
"KAYZEN_BASIC_AUTH": "auth token"
}
}
}
}Node.js (v16 以上)
npm (v7以上)
Kayzen API 認証情報
# Install dependencies
npm install
# Build the project
npm run build
# Start the server
npm start
# Development mode with hot-reload
npm run devkayzen-mcp/
├── src/
│ ├── server.ts # MCP server implementation
│ └── kayzen-client.ts # Kayzen API client
├── dist/ # Compiled JavaScript
└── package.json # Project configuration主な依存関係:
@modelcontextprotocol/sdk : ^1.7.0
axios :^1.8.3
dotenv :^16.4.7
zod :^3.24.2
サーバーはさまざまなエラー シナリオを処理します。
認証失敗
無効なAPIリクエスト
ネットワークの問題
トークンの有効期限と更新
無効なパラメータ
MITライセンス
フィードモブ
| Name | Required | Description | Default |
|---|---|---|---|
| end_date | No | End date in YYYY-MM-DD format | |
| report_id | Yes | ID of the report to fetch results for | |
| start_date | No | Start date in YYYY-MM-DD format |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0get_report_resultslist_reportsScored across 2 tools
The two tools have clearly distinct purposes: 'get_report_results' appears to retrieve specific report data, while 'list_reports' enumerates available reports. There is no overlap in functionality, making it easy for an agent to choose the correct tool based on whether it needs to list reports or fetch results from a particular report.
Both tools follow a consistent verb_noun naming pattern ('get_report_results' and 'list_reports'), using snake_case throughout. The verbs 'get' and 'list' are standard and predictable, making the tool names easy to understand and use without confusion.
With only two tools, the server feels under-scoped for an analytics domain, which typically involves operations like creating, updating, filtering, or deleting reports. This minimal set may force agents to work around gaps, such as being unable to generate new reports or modify existing ones, limiting functionality.
The tool surface is severely incomplete for analytics: it allows listing and retrieving reports but lacks essential operations like creating, updating, deleting, or filtering reports. This creates significant gaps that will likely cause agent failures when trying to perform common analytics tasks beyond basic read-only access.
Access Google & Meta Ads data via AI. Analyse campaign performance in seconds.
Conversational access to advertising performance data, creative analysis, and campaign insights
Conversational access to advertising performance data, creative analysis, and campaign insights
Marketing intelligence API for AI agents. Real campaign data, not LLM guesses.
