About ServerSearcher
ServerSearcher is an MCP server for finding and comparing dedicated, VPS, GPU, and cloud server plans. It helps people choosing infrastructure filter a catalogue using specifications, location, price, and provider details.
Key features
- 01Search plans by hardware specifications
- 02Filter by server type, location, and price
- 03Find GPU models and provider capabilities
- 04View provider profiles and locations
- 05Compare two providers side by side
Use cases
- 01Find a dedicated server matching a RAM and budget limit
- 02Locate GPU or cloud plans in a chosen city
- 03Review provider locations and available plans
- 04Compare two hosting providers
MCP support
Exposes search_servers, get_provider_info, and compare_providers to MCP clients over Streamable HTTP. The public endpoint requires no authentication.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://www.serversearcher.com/api/mcpclaude mcp add --transport http com-serversearcher-serversearcher https://www.serversearcher.com/api/mcp{"mcpServers":{"com-serversearcher-serversearcher":{"type":"http","url":"https://www.serversearcher.com/api/mcp"}}}Open in Cursor{"servers":{"com-serversearcher-serversearcher":{"type":"http","url":"https://www.serversearcher.com/api/mcp"}}}Open in VS Code{"mcpServers":{"com-serversearcher-serversearcher":{"type":"http","url":"https://www.serversearcher.com/api/mcp"}}}{"mcpServers":{"com-serversearcher-serversearcher":{"serverUrl":"https://www.serversearcher.com/api/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
3| Parameter | Type | Required | Description |
|---|---|---|---|
| page | number | No | Page number, 1-based |
| type | array | No | Plan types to include. Omit for all types. |
| minRam | number | No | Minimum RAM in GB |
| cpuModel | string | No | Substring of the CPU model, e.g. "EPYC". `*` and `?` are glob wildcards: "EPYC*9004" matches "AMD EPYC 9004". |
| currency | array | No | ISO 4217 currency codes to restrict prices to, e.g. ["USD","EUR"] |
| gpuModel | string | No | Exact GPU model name, e.g. "NVIDIA H100" |
| location | array | No | Locations: an ISO country code ("DE"), a country and city ("DE-frankfurt"), or a city id |
| maxPrice | number | No | Maximum listed monthly price, in the listed currency |
| pageSize | number | No | Results per page, maximum 150 |
| provider | array | No | Provider slugs (e.g. "runpod") or internal provider ids — slugs are what URLs and get_provider_info use and are resolved for you. |
| diskTypes | array | No | Disk types, e.g. NVMe, SSD, HDD |
| unmetered | boolean | No | Only plans with unmetered bandwidth (no monthly transfer quota) |
| sortConfig | enum | No | Sort order. The pricePer* keys rank by USD price per unit of RAM or CPU.priceAsc | priceDesc | ramAsc | ramDesc | cpuAsc | cpuDesc | diskDesc | bandwidthDesc | pricePerRam | pricePerCpu | pricePerRamCpu |
| minCpuCores | number | No | Minimum CPU cores |
| minGpuCount | number | No | Minimum number of GPUs |
| hasTerraform | boolean | No | Only providers with an official or community Terraform provider |
| minBandwidth | number | No | Minimum included bandwidth in TB per month |
| minPortSpeed | number | No | Minimum network port speed in Mbps; matches either port |
| cloudServices | array | No | Restrict to providers offering any of these managed services, e.g. ["OBJECT_STORAGE","MANAGED_KUBERNETES"] |
| minCpuThreads | number | No | Minimum CPU threads |
| minCpuFrequency | number | No | Minimum CPU base frequency in GHz |
| minDiskCapacity | number | No | Minimum capacity of a single disk, in GB |
Details
From the project READMEServerSearcher MCP
Public Model Context Protocol endpoint for serversearcher.com — a catalogue of dedicated, VPS, GPU and cloud server plans with live pricing, stock and locations.
Add it to any MCP client and ask questions like "find me a dedicated server with 64 GB RAM in Germany under €60" straight from your agent.
Endpoint
| URL | https://www.serversearcher.com/api/mcp |
| Transport | Streamable HTTP (JSON-RPC 2.0); POST for requests, GET/OPTIONS supported |
| Protocol | MCP 2024-11-05 |
| Auth | None — the endpoint is public |
| Discovery | /.well-known/mcp.json |
| Alias | https://www.serversearcher.com/mcp |
Rate-limited politely; bursts get a 429 with a Retry-After.
Tools
| Tool | What it does |
|---|---|
search_servers | Search the plan catalogue by spec — type (dedicated/virtual/cloud/gpu), min RAM/CPU/disk/bandwidth, GPU model and count, price ceiling, currency, location (ISO country, COUNTRY-city, or city id), and provider capabilities (hasTerraform, cloud services). Returns plans with on-site URLs and checkout links. |
get_provider_info | Provider profile, plan counts, locations and on-site URLs from a provider slug (e.g. hetzner-cloud). |
compare_providers | Side-by-side comparison of two providers by slug, with the on-site comparison URL. |
Client configuration
Claude Desktop (claude_desktop_config.json):
{
"mcpServers": {
"serversearcher": {
"type": "streamable-http",
"url": "https://www.serversearcher.com/api/mcp"
}
}
}
Any client that reads server configs:
{
"mcpServers": {
"serversearcher": {
"type": "streamable-http",
"url": "https://www.serversearcher.com/mcp"
}
}
}
Quick check
curl -s https://www.serversearcher.com/api/mcp \
-H 'content-type: application/json' \
-H 'accept: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'
curl -s https://www.serversearcher.com/api/mcp \
-H 'content-type: application/json' \
-H 'accept: application/json' \
-d '{"jsonrpc":"2.0","id":2,"method":"tools/call",
"params":{"name":"search_servers",
"arguments":{"type":["dedicated"],"minRam":64,"maxPrice":60}}}'
Links
- Site: https://www.serversearcher.com
- Machine-readable site map for agents: https://www.serversearcher.com/llms.txt
License
MIT — see LICENSE.
Comments
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.