Agent API

Structured discovery surfaces for AI agents.

These endpoints are built for plan comparison, domain search, and checkout preparation. State-changing actions stay behind secure approval and checkout controls.

Catalog JSONDomain searchMCP JSON-RPC
Quick start

Use the public surfaces in the right order.

The safest production setup is discovery first, validation second, and verified checkout handoff last.

Start with discovery

Use public pages, the catalog API, or MCP tools to compare plans and search domains before a customer chooses.

Prepare the cart safely

Validate domain requirements, billing choices, and checkout readiness without writing directly to the billing platform.

Keep writes verified

Order creation, payment, and account changes should stay behind verified customer handoff and secure checkout controls.

Endpoints

Current public endpoints.

Each surface below is safe for product discovery and domain planning without exposing direct write actions.

GET/api/agent/catalog

Returns public product categories, plans, pricing, currencies, and approval policy.

  • Best for structured product discovery.
  • Keeps checkout entry URLs alongside each plan.
  • Read-only by design.
GET/api/agent/domain-search?query=example&tld=.com

Checks public domain availability for a brand term or exact domain without mutating cart state.

  • Use for domain suggestions and availability checks.
  • Supports exact domains and brand-style queries.
  • Returns alternatives when the exact match is unavailable.
POST/api/mcp

JSON-RPC MCP endpoint exposing list_products, search_domains, and checkout_requirements tools.

  • Use for agent runtimes that support MCP or MCP-style tool calls.
  • Returns structured tool lists and tool outputs.
  • Direct write actions are not exposed.
Code samples

Reference requests you can wire into an agent today.

Use these as production-ready starting points for curl, browser fetch calls, and JSON-RPC tool execution.

Catalog with curl
curl -s https://onenetservers.net/api/agent/catalog
Domain search with fetch
const response = await fetch("https://onenetservers.net/api/agent/domain-search?query=samplebrand&tld=.com");
const data = await response.json();
MCP tool call
POST https://onenetservers.net/api/mcp
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "search_domains",
    "arguments": {
      "query": "samplebrand",
      "tlds": [".com", ".ng"]
    }
  }
}
Production checklist

The minimum bar for shipping an agent against these APIs.

These checks keep the integration useful for customers without turning discovery tooling into an unsafe billing shortcut.

Treat all discovery surfaces as read-only.
Require customer approval before any checkout or account mutation.
Validate domain requirements before collecting billing details.
Handle temporary billing-platform failures with generic user-safe messages.
Use MCP for tool calls, not for hidden order creation.