Cloud Services
cloudfinops.ru
MCP server for discovering cloud providers, searching and comparing cloud products, and estimating monthly compute or GPU costs.
ENDPOINT 1
https://cloudfinops.ru/mcp
MCP server metadata
- Name
- cloudfinops
- Version
- 1.0.0
CloudFinOps provides read-only public cloud pricing and configuration estimates for Russia through REST and MCP. It does not access user cloud accounts or provision resources. Use list_providers to discover supported provider IDs. Use search_products for SKU discovery, get_product for detailed price rules and sources, list_product_alternatives for comparable SKUs, and create_estimate for complete monthly VM/GPU costs. Do not replace a configuration estimate with the price of one CPU or GPU billing component. For AI token prices, call search_products with services=["ai"], units=["token"] and no q. Add modelIds, tokenDirections or meters for exact selection. categories=["ai"] alone also includes ML infrastructure and per-request services. Use Price.unitQuantity when comparing token packs; a token SKU does not represent a GPU rental. For exact GPU models, CPU/RAM sizes and other characteristics, use attributes filters with eq, in or inclusive range; GET /attributes lists fields per category and observed values. The search_products input schema describes their types and categories. Null never matches. Numeric SKU metadata is not a request to assemble a configuration. Use the documentation resource at https://cloudfinops.ru/llms-full.txt and the OpenAPI resource at https://cloudfinops.ru/api/v1/openapi.json for parameters and examples. Both are available through resources/read. Money is represented as decimal strings. Respect unitQuantity, unit, tiers and VAT; null is not zero. Compare only priced estimates or alternatives with priceComparable=true. Report matchedResource, differences and the 720-hour assumption. A catalog status is not live capacity confirmation. Cite CloudFinOps and the underlying product source with checkedAt; catalogVersion identifies a snapshot, not the date a provider last changed its price. For missing data, explain the gap rather than inventing a price.
Known tools 5
list_providersDiscover cloud providers, their IDs, catalog categories, product counts and public pricing sources.
Inferred read-onlysearch_productsDiscover characteristics with GET /attributes, then set attributes to an object of eq, in or inclusive range filters.
Inferred read-onlylist_product_alternativesFind alternatives at other providers for a product ID returned by search_products.
Inferred read-onlycreate_estimateCalculate a monthly cost estimate from public Russian cloud tariffs for compute (vcpu and memoryGiB) or GPU (gpuModel and gpuCount).
Inferred read-onlyCONNECT WITH APPROVAL
Client installation
Review this server and its permissions before adding it. Secret placeholders must be set locally.
Codex
~/.codex/config.toml
[mcp_servers.cloudfinops]
url = "https://cloudfinops.ru/mcp"
enabled = true
Claude Code
.mcp.json
{
"mcpServers": {
"cloudfinops": {
"type": "http",
"url": "https://cloudfinops.ru/mcp"
}
}
}
Claude Desktop
Settings → Connectors → Add custom connector
Name: cloudfinops
Remote MCP URL: https://cloudfinops.ru/mcp
Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.
Cursor
.cursor/mcp.json
{
"mcpServers": {
"cloudfinops": {
"url": "https://cloudfinops.ru/mcp"
}
}
}
Visual Studio Code
.vscode/mcp.json
Add to Visual Studio Code{
"servers": {
"cloudfinops": {
"type": "http",
"url": "https://cloudfinops.ru/mcp"
}
}
}
Generic MCP
Client-specific MCP configuration
{
"name": "cloudfinops",
"transport": "streamable-http",
"url": "https://cloudfinops.ru/mcp"
}
MCP Inspector
Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.
TRUST AND VERIFICATION EVIDENCE
Loading Trust v2 evidence…
Checking the associated registrable domain. The BuiltWith key remains server-side.
Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.