AIエージェント向けのクロスクラウドインフラ可視化ツール。AWS、Vercel、GCP、Cloudflareにまたがる問題を、エディタから離れることなく診断します。
課題 | CloudPulseによる解決策 |
Vercelでフロントエンドエラー発生 → AWSコンソールを開く必要がある |
|
AIがセキュリティグループでポート5432がブロックされているか確認できない |
|
Lambdaの同時実行制限に気づかず到達してしまう |
|
デバッグ前にトポロジーが不明 |
|
Related MCP server: Daemoon
npx cloudpulse-mcpサーバーは、マシン上に既に存在する認証情報（AWS CLI、環境変数など）を自動的に検出します。
Claude Desktop – ~/Library/Application Support/Claude/claude_desktop_config.json に追加:
{
"mcpServers": {
"cloudpulse": {
"command": "npx",
"args": ["-y", "cloudpulse-mcp"],
"env": {
"VERCEL_TOKEN": "<your-vercel-token>",
"AWS_PROFILE": "default",
"AWS_REGION": "us-east-1"
}
}
}
}Cursor – プロジェクト内の .cursor/mcp.json に追加:
{
"mcpServers": {
"cloudpulse": {
"command": "npx",
"args": ["-y", "cloudpulse-mcp"],
"env": {
"VERCEL_TOKEN": "<your-vercel-token>",
"AWS_REGION": "us-east-1"
}
}
}
}VS Code + GitHub Copilot (エージェントモード) – VS Code 1.99以降とGitHub Copilot拡張機能が必要です。
まず、プロジェクトをビルドします:
npm run build次に、このリポジトリ内に .vscode/mcp.json を作成します:
{
"servers": {
"cloudpulse": {
"type": "stdio",
"command": "node",
"args": ["${workspaceFolder}/dist/index.js"],
"env": {
"VERCEL_TOKEN": "${env:VERCEL_TOKEN}",
"AWS_REGION": "${env:AWS_REGION}",
"AWS_PROFILE": "${env:AWS_PROFILE}"
}
}
}
}${env:VAR} はシェル環境から読み込まれるため、ソース管理にシークレットが含まれることはありません。
使用方法: Copilot Chat を開き、エージェントモードに切り替え、Select Tools をクリックしてCloudPulseツールを有効にし、自然言語で質問してください:
Why can't my Vercel project reach AWS RDS instance "my-db"?CloudPulseは読み取り専用・非保存ポリシーに従います:
認証情報 | 提供方法 |
AWS |
|
Vercel |
|
Vercel Team |
|
GCP |
|
Cloudflare |
|
認証情報はログに記録されたり保存されたりすることはありません。 すべての値は呼び出し時に環境変数から読み取られます。
list_cloud_topology設定されたすべてのプラットフォームをスキャンし、統合されたサービスマップを返します。
Input (all optional):
platforms – ["aws", "vercel"] filter platforms
aws_region – "us-east-1"get_correlated_logsVercelとAWS CloudWatchのログを取得し、1つのタイムラインにマージします。
Input:
start_time * – ISO-8601 or epoch ms e.g. "2024-06-01T10:00:00Z"
end_time – defaults to now
trace_id – filter by trace/request ID across all sources
aws_log_group_prefix – default "/aws/lambda"
vercel_project – project name or ID
aws_regiondiagnose_service_linkサービスAがリソースBに到達できない理由を調査します。
Input:
source_service * – "vercel" | "lambda" | "ec2" | ...
target_resource * – "<type>:<id>" e.g. "aws-rds:my-db", "external-api:https://..."
port – auto-detected (5432 for RDS, 443 for APIs, ...)
vercel_project
aws_region実行されるチェック:
Vercel環境変数に DATABASE_URL / DB_URL が含まれているか
AWSセキュリティグループが必要なポートでのインバウンドTCPを許可しているか
外部APIのHEAD到達可能性テスト
check_resource_limitsクォータを照会し、制限に近づいているリソースにフラグを立てます。
Input (all optional):
platforms – filter platforms
warn_threshold – usage % to warn at (default 80)
aws_regionフェーズ | ステータス | スコープ |
1 – MVP | ✅ 完了 | Vercel + AWS (Lambda, RDS, CloudWatch, Security Groups, S3) |
2 – 拡張 | ✅ 完了 | GCP Cloud Run + Cloud SQL + Logging; Cloudflare Workers + Pages; S3 CORS |
3 – インテリジェンス | 🔜 | CORS、504タイムアウト、コールドスタートループ向けの事前構築済み診断プレイブック |
git clone https://github.com/Galadriel-Tech-Solutions/cloudpulse-mcp
cd cloudpulse-mcp
npm install
npm run dev # run from source with tsx
npm run build # compile to dist/src/
├── index.ts # MCP server + tool registration
├── types.ts # shared domain types
├── utils.ts # concurrency, formatting helpers
├── providers/
│ ├── aws/
│ │ ├── index.ts # client factory + isAWSConfigured()
│ │ ├── cloudwatch.ts # CloudWatch Logs
│ │ ├── lambda.ts # Lambda function listing
│ │ ├── rds.ts # RDS/Aurora instances & clusters
│ │ ├── ec2.ts # Security Group inspection
│ │ ├── s3.ts # S3 buckets + CORS checks
│ │ └── quotas.ts # Service Quotas API
│ ├── gcp/
│ │ ├── index.ts # isGCPConfigured() + resolveGCPProject()
│ │ ├── cloud-run.ts # Cloud Run services
│ │ ├── cloud-sql.ts # Cloud SQL instances (sqladmin v1beta4)
│ │ └── logging.ts # Cloud Logging
│ ├── cloudflare/
│ │ └── index.ts # Pages, Workers, Worker tail logs (WebSocket)
│ └── vercel/
│ └── index.ts # Vercel REST API v9
└── tools/
├── list-cloud-topology.ts
├── get-correlated-logs.ts
├── diagnose-service-link.ts
└── check-resource-limits.tssrc/providers/<platform>/index.ts を作成し、以下をエクスポートします:
is<Platform>Configured(): boolean
プロバイダー固有のデータ関数
src/tools/ 配下の関連ツールにその関数を組み込みます
src/types.ts の CloudPlatform ユニオン型にプラットフォーム名を追加します
MIT © CloudPulse Contributors
Query quota limits and current usage across configured cloud platforms. Highlights resources approaching or exceeding their limits (default warning threshold: 80%). Use this to proactively catch Lambda concurrency limits, Vercel plan caps, and similar issues before they cause outages.
Inputs: platforms, warn_threshold, aws_region.
Diagnose connectivity issues between a source service and a target resource. Checks Vercel environment variables, AWS Security Group inbound rules on the required port, and external API reachability. Returns a list of diagnostic results with actionable messages.
Inputs: source_service (required), target_resource (required), port, aws_region, +2 more.
Scan all configured cloud platforms (AWS, Vercel, GCP, Cloudflare) and return a unified topology of active services including their endpoints and regions. Run this first to understand the infrastructure landscape.
Inputs: platforms, aws_region.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 4 tools
Each tool has a clearly distinct purpose with no overlap: check_resource_limits focuses on quota monitoring, diagnose_service_link on connectivity diagnostics, get_correlated_logs on log aggregation, and list_cloud_topology on infrastructure discovery. The descriptions reinforce unique scopes, making misselection unlikely.
All tool names follow a consistent verb_noun pattern (e.g., check_resource_limits, diagnose_service_link), using snake_case uniformly. This predictability aids agent understanding and tool selection without confusion.
Four tools is a reasonable count for a cloud monitoring server, covering key areas like limits, diagnostics, logs, and topology. It feels slightly lean but well-scoped, as each tool addresses a distinct monitoring need without bloat.
The toolset covers core cloud monitoring workflows: proactive limits checking, connectivity diagnosis, log correlation, and topology mapping. Minor gaps exist, such as lack of alerting or remediation tools, but agents can work around these with the provided diagnostic and data-fetching capabilities.
Errors, logs and AI-call costs for apps on Node.js, Next.js, Cloudflare Workers or the browser. Ask your AI what’s broken, why, and how to fix it, right inside Cursor or Claude.
Open-source agent that observes and fixes your application. Query logs, traces, metrics, incidents.
AI agent observability for production traces, natural-language insights, and improvement loops.
Data + AI observability — monitor and troubleshoot production-grade agents and the context they use.
