mcpField
C
CloudScope MCP

CloudScope MCP

Local
github.com/alexpotaCloud & Infra1Updated 22w ago

About CloudScope MCP

CloudScope MCP provides read-only cloud cost analysis for teams using Azure or GCP. Its tools and guided prompts cover spending, budgets, anomalies, forecasts, idle resources, and optimization recommendations.

Key features

  1. 01Break down costs by service, tag, or resource
  2. 02Compare spending periods and detect anomalies
  3. 03Check budgets and forecast costs
  4. 04Find idle and untagged resources
  5. 05Get cost optimization recommendations
  6. 06Generate guided cost reports

Use cases

  1. 01Review monthly cloud spending and changes
  2. 02Investigate a recent cost spike
  3. 03Find idle resources and potential savings
  4. 04Prepare cost allocation reports by tag

MCP support

Exposes cloud cost analysis and monitoring tools, along with guided report prompts, to MCP clients. Azure and GCP are supported; access is read-only.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"cloudscope": {
"command": "npx",
"args": [
"-y",
"cloudscope-mcp"
],
"env": {
"AZURE_SUBSCRIPTION_ID": "<AZURE_SUBSCRIPTION_ID>",
"AZURE_TENANT_ID": "<AZURE_TENANT_ID>",
"AZURE_CLIENT_ID": "<AZURE_CLIENT_ID>",
"AZURE_CLIENT_SECRET": "<AZURE_CLIENT_SECRET>",
"GOOGLE_CLOUD_PROJECT": "<GOOGLE_CLOUD_PROJECT>",
"GCP_BILLING_TABLE": "<GCP_BILLING_TABLE>",
"GCP_PROJECT_ID": "<GCP_PROJECT_ID>",
"GCP_BILLING_ACCOUNT_ID": "<GCP_BILLING_ACCOUNT_ID>",
"GOOGLE_APPLICATION_CREDENTIALS": "<GOOGLE_APPLICATION_CREDENTIALS>"
}
}
}
}
Run commandnpx -y cloudscope-mcp
Environment variables to set
AZURE_SUBSCRIPTION_IDAzure subscription ID to query for cost data. Optional; auto-detected from az login session if not set.
AZURE_TENANT_IDAzure tenant ID for service principal authentication. Optional; falls back to DefaultAzureCredential (e.g. az login) if unset.
AZURE_CLIENT_IDService principal client ID. Optional; required only when using ClientSecretCredential.
AZURE_CLIENT_SECRETSecretService principal client secret. Optional; required only when using ClientSecretCredential.
GOOGLE_CLOUD_PROJECTGCP project ID. Recommended to set explicitly in the MCP client env block — may not be inherited from gcloud shell configuration.
GCP_BILLING_TABLEFully-qualified BigQuery billing export table (project.dataset.table_name). Required for GCP cost queries.
GCP_PROJECT_IDOverride GCP project ID if the billing dataset lives in a different project than GOOGLE_CLOUD_PROJECT.
GCP_BILLING_ACCOUNT_IDGCP billing account ID (format: XXXXXX-XXXXXX-XXXXXX). Optional; needed only for budget monitoring.
GOOGLE_APPLICATION_CREDENTIALSSecretPath to GCP service account JSON key file. Optional; falls back to Application Default Credentials (gcloud auth).

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmcloudscope-mcp0.3.3

Tools

15

Saved tool list · updated 3w ago

ParameterTypeRequiredDescription
providerenumNoCloud provider to query (azure or gcp)azure | gcp
start_datestringNoStart date (YYYY-MM-DD). Defaults to first of current month.
end_datestringNoEnd date (YYYY-MM-DD). Defaults to today.
group_byenumNoHow to group costs: service, resource_group, tag, or regionservice | resource_group | tag | region

Details

From the project README

CloudScope MCP

Ask your AI about your cloud bill.

npm version MCP Registry Node.js TypeScript License: MIT

What you get

Ask once: "Give me this month's cost review" — and CloudScope runs the full workflow (spend, deltas, anomalies, idle resources, budgets, forecast) and hands back a decision-ready report, not raw numbers to interpret yourself:

TL;DR

Month-to-date Azure spend is $4,312 (May 1–29), tracking +11% vs April. The mover is App Service (+$280) — a new staging slot left running. 2 idle resources cost $94/month; prod-budget is at 78% with 2 days left. Top action: delete the orphaned managed disk → save $63/month.

Spend This Month

ServiceCost%
Virtual Machines$2,14049.6%
App Service$89020.6%
Storage$61214.2%
Redis Cache$4109.5%
Networking$2606.0%

vs Last Month

+$427 (+11%) overall. App Service +$280 drove most of it; everything else flat. → new staging-slot-2, created May 12, never torn down.

Idle & Wasted Resources

ResourceType$/month
orphaned-osdisk-01Managed Disk$63
old-lb-ipPublic IP$31

Budget Status

prod-budget — 78% used, $4,312 / $5,500, on track (2 days left).

Recommended Actions

  1. Delete orphaned-osdisk-01 (unattached 40 days) → $63/month
  2. Tear down staging-slot-2 if staging is idle → ~$280/month
  3. Release old-lb-ip (no binding) → $31/month

Illustrative example showing the report format. Run /cloudscope:monthly-cost-review against your own subscription for live numbers.

Install

Claude CodeCursorVS Code
claude mcp add cloudscope -- npx -y cloudscope-mcpInstallInstall

Azure is auto-detected from your az login session. GCP requires BigQuery billing export setup (see below).

Supported Providers

ProviderStatus
Azure✅ Supported
GCP✅ Supported
AWSUse AWS's official server

Prerequisites

Azure
  • Azure CLI installed and logged in (az login)
  • Cost Management Reader role on the subscription
GCP
  • gcloud CLI installed and logged in (gcloud auth application-default login)
  • Billing export to BigQuery enabled — this is a one-time manual setup step, not enabled by default (the detailed export is recommended for resource-level cost queries)
  • BigQuery Data Viewer + BigQuery Job User roles on the dataset project
  • Note: BigQuery on-demand queries cost $6.25 per TiB scanned (pricing). The first 1 TiB/month is free. Billing export tables are small — typical CloudScope queries cost <$0.01 each.

Configuration

Azure (zero-config)

CloudScope auto-detects your subscription from az login. Just add the server:

{
  "mcpServers": {
    "cloudscope": {
      "command": "npx",
      "args": ["-y", "cloudscope-mcp"]
    }
  }
}

To target a specific subscription, add an env block:

{
  "mcpServers": {
    "cloudscope": {
      "command": "npx",
      "args": ["-y", "cloudscope-mcp"],
      "env": {
        "AZURE_SUBSCRIPTION_ID": "your-subscription-id"
      }
    }
  }
}
Advanced: Service Principal (CI/CD & automated environments)
VariableDescription
AZURE_TENANT_IDAzure AD tenant ID
AZURE_CLIENT_IDApp registration client ID
AZURE_CLIENT_SECRETApp registration client secret

Set these alongside AZURE_SUBSCRIPTION_ID in the env block above.

GCP

GCP requires a BigQuery billing export table. Find your table name in GCP Console > Billing > Billing export > BigQuery export.

{
  "mcpServers": {
    "cloudscope": {
      "command": "npx",
      "args": ["-y", "cloudscope-mcp"],
      "env": {
        "GOOGLE_CLOUD_PROJECT": "my-project",
        "GCP_BILLING_TABLE": "my-project.my_dataset.gcp_billing_export_resource_v1_XXXXXX"
      }
    }
  }
}

Set GOOGLE_CLOUD_PROJECT to your GCP project ID. Override with GCP_PROJECT_ID if your billing dataset lives in a different project.

These variables apply only when using GCP; Azure-only users can skip them.

VariableDescriptionRequired
GOOGLE_CLOUD_PROJECTGCP project IDYes
GCP_BILLING_TABLEFully-qualified BigQuery table (project.dataset.table)Yes
GCP_PROJECT_IDOverride project ID if different from GOOGLE_CLOUD_PROJECTNo
GCP_BILLING_ACCOUNT_IDBilling account ID for budget monitoringNo
GOOGLE_APPLICATION_CREDENTIALSPath to service account JSON key fileNo
Both Providers
{
  "mcpServers": {
    "cloudscope": {
      "command": "npx",
      "args": ["-y", "cloudscope-mcp"],
      "env": {
        "AZURE_SUBSCRIPTION_ID": "your-subscription-id",
        "GOOGLE_CLOUD_PROJECT": "my-project",
        "GCP_BILLING_TABLE": "my-project.my_dataset.gcp_billing_export_resource_v1_XXXXXX"
      }
    }
  }
}

Azure is auto-detected from az login. Add AZURE_SUBSCRIPTION_ID to target a specific subscription.

Example Questions

  • "How much did Azure cost last month?"
  • "Show GCP spending by service for the last 7 days"
  • "Any cost anomalies this week on GCP?"
  • "What will Azure cost next month?"
  • "Show me cost optimization recommendations for GCP"
  • "Compare Azure and GCP costs across all projects and subscriptions"

Security

CloudScope is read-only. It cannot create, modify, or delete any cloud resources. Azure uses Cost Management Reader permissions. GCP uses BigQuery Data Viewer + Job User with no write access.

FAQ

Does this modify my cloud resources? No. Read-only access only.

Do I need a service principal? No. az login (Azure) or gcloud auth application-default login (GCP) works for local use.

Does the Azure Cost Management API cost money? No. It's free.

Do GCP BigQuery cost queries cost money? Yes, but typically <$0.01 per query ($6.25/TiB scanned, first 1 TiB/month free). See BigQuery pricing.

Can I use both Azure and GCP at the same time? Yes. Configure both sets of env vars and CloudScope queries whichever provider you specify in each tool call.

Development

git clone https://github.com/alexpota/cloudscope-mcp.git
cd cloudscope-mcp
npm install
npm run build
npm test
npx @modelcontextprotocol/inspector node dist/index.js

Reference

Tools

Most tools accept a provider parameter (azure or gcp); the provider-specific tools (list_subscriptions, list_projects, and the cross-account tools get_cross_subscription_costs / get_cross_project_costs) are fixed to their own cloud. The default provider is auto-detected based on which providers are configured.

Cost Analysis

ToolDescriptionKey Parameters
get_cost_summarySpending breakdown by service, group, or regionstart_date, end_date, group_by
get_cost_by_tagCosts grouped by a tag/label keytag_key, start_date, end_date
compare_periodsSide-by-side cost comparison of two date rangesperiod_a_start/end, period_b_start/end
top_spending_resourcesMost expensive individual resourcesdays, limit
get_cross_subscription_costsCombined costs across Azure subscriptionssubscription_ids, start_date, end_date
get_cross_project_costsCombined costs across GCP projectsproject_ids, start_date, end_date

Monitoring

ToolDescriptionKey Parameters
detect_anomaliesFind spending spikes vs previous perioddays, threshold
check_budgetsBudget status, current spend, projected overage(none)
get_cost_forecastPredict spending based on current trendsdays

Optimization

ToolDescriptionKey Parameters
list_recommendationsCost optimization suggestions (Azure Advisor / GCP Recommender)category
find_idle_resourcesProvisioned but unused resources with cost estimates(none)
find_untagged_resourcesResources with no tags/labels (cost attribution gaps)(none)

Utility

ToolDescriptionKey Parameters
get_current_dateToday's date and current/previous month bounds(none)
list_subscriptionsAzure subscriptions with active indicator(none)
list_projectsGCP projects with active indicator(none)
Prompts

Guided workflows that produce structured reports. All prompts accept an optional provider argument (azure or gcp). In Claude Code, type /cloudscope: to see all prompts. In Claude Desktop, click the + button → Connectors → cloudscope.

PromptDescriptionArguments
monthly-cost-reviewComplete monthly review: spending, last-month comparison, anomalies, top resources, budgets, forecast, savings opportunitiesprovider (optional)
waste-auditFind wasted spend: top resources, optimization recommendations, at-risk budgets, total potential savingsprovider (optional)
cost-spike-investigationRoot-cause analysis for a cost increase: which services, which resources, trend vs one-time, recommended actionsdays (optional), provider (optional)
executive-summaryBrief non-technical cost summary for leadership: spend, trend, budget status, top drivers, forecast, key recommendationprovider (optional)
chargeback-reportCost allocation by tag/label key for chargeback: spending per value, untagged resources, tagged vs untagged split, month-over-monthtag_key (required), provider (optional)

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.