StableDomains

Install StableDomains in your agent

MCP endpoint https://stabledomains.net/api/mcp (Streamable HTTP), registry name net.stabledomains/registrar. Search, buy with USDC/USDT, manage DNS and renew from any MCP client.

New here? Start with the connect your agent guide (four steps, client how-tos, prompts and the safety rules). This page is the full install list.

One-click MCP install

Full MCP endpoint https://stabledomains.net/api/mcp (search, buy, DNS). Directories and read-only chat apps can use https://stabledomains.net/api/mcp/discover instead — same search and quotes, no purchase tools, a checkout link when someone wants to buy. Search works without a key. To buy or manage domains, sign in with OAuth when your client asks, or add Authorization: Bearer sd_live_... (free key: curl -s -X POST https://stabledomains.net/api/v1/agents/signup).

Cursor
~/.cursor/mcp.json or .cursor/mcp.json
Add to Cursor

Official Cursor install link. Opens Cursor and prompts to add the full MCP server (search, buy, DNS). Search works immediately; sign in with OAuth or an API key when you need to buy.

{
  "mcpServers": {
    "stabledomains": {
      "url": "https://stabledomains.net/api/mcp"
    }
  }
}

# with an API key:
{
  "mcpServers": {
    "stabledomains": {
      "url": "https://stabledomains.net/api/mcp",
      "headers": {
        "Authorization": "Bearer sd_live_..."
      }
    }
  }
}

Claude setup

claude mcp add --transport http stabledomains https://stabledomains.net/api/mcp
# with an API key:
claude mcp add --transport http stabledomains https://stabledomains.net/api/mcp --header "Authorization: Bearer sd_live_..."

Claude.ai / Desktop: Settings → Connectors → Add custom connector → paste https://stabledomains.net/api/mcp and sign in.

Major agents

Cursor

The most used AI code editor; one click adds every tool.

Click the button, then approve in Cursor. For buying, add a Bearer key.

One-click install
{ "mcpServers": { "stabledomains": { "url": "https://stabledomains.net/api/mcp", "headers": { "Authorization": "Bearer <key>" } } } }
Cursor docs →

Claude (Code, Desktop, claude.ai)

Anthropic's apps, where MCP started. OAuth sign-in built in.

Claude Code command below. Claude.ai / Desktop: Settings → Connectors → Add custom connector → paste the URL and sign in.

claude mcp add --transport http stabledomains https://stabledomains.net/api/mcp --header "Authorization: Bearer <key>"
Claude (Code, Desktop, claude.ai) docs →

ChatGPT

The largest chat assistant; custom MCP connectors in developer mode.

Settings → Security and login → turn on Developer mode. Then Plugins → + → Add custom MCP server → paste the URL (OAuth or no auth).

https://stabledomains.net/api/mcp
ChatGPT docs →

VS Code + GitHub Copilot

Copilot agent mode runs MCP tools in the most popular editor.

Click Install in VS Code, or add to .vscode/mcp.json.

One-click install
{ "servers": { "stabledomains": { "type": "http", "url": "https://stabledomains.net/api/mcp", "headers": { "Authorization": "Bearer <key>" } } } }
VS Code + GitHub Copilot docs →

OpenAI Codex CLI

OpenAI's coding agent for the terminal.

One command:

codex mcp add stabledomains --url https://stabledomains.net/api/mcp
OpenAI Codex CLI docs →

Gemini CLI

Google's open-source terminal agent.

One command:

gemini mcp add --transport http stabledomains https://stabledomains.net/api/mcp
Gemini CLI docs →

Windsurf

Cascade agent with first-class MCP support.

Add to ~/.codeium/windsurf/mcp_config.json:

{ "mcpServers": { "stabledomains": { "serverUrl": "https://stabledomains.net/api/mcp" } } }
Windsurf docs →

Cline

Open-source autonomous coding agent for VS Code.

MCP Servers → Configure → cline_mcp_settings.json. Set type explicitly.

{ "mcpServers": { "stabledomains": { "type": "streamableHttp", "url": "https://stabledomains.net/api/mcp" } } }
Cline docs →

Zed

Fast native editor with an agent panel.

Add to Zed settings.json:

{ "context_servers": { "stabledomains": { "url": "https://stabledomains.net/api/mcp" } } }
Zed docs →

Continue

Open-source assistant for VS Code and JetBrains.

Add .continue/mcpServers/stabledomains.yaml:

name: StableDomains
version: 0.0.1
schema: v1
mcpServers:
  - name: stabledomains
    type: streamable-http
    url: https://stabledomains.net/api/mcp
Continue docs →

OpenAI Agents SDK

Build your own agent in Python with hosted MCP tools.

Python:

from agents import Agent
from agents.mcp import MCPServerStreamableHttp
async with MCPServerStreamableHttp(params={"url": "https://stabledomains.net/api/mcp", "headers": {"Authorization": "Bearer sd_live_..."}}) as sd:
    agent = Agent(name="domains", mcp_servers=[sd])
OpenAI Agents SDK docs →

Grok / xAI API

xAI's API can call remote MCP servers as tools.

Pass the URL as a remote MCP tool in your xAI API request.

https://stabledomains.net/api/mcp
Grok / xAI API docs →

Emerging agents and frameworks

Goose

Block's open-source desktop and CLI agent.

goose configure → Add Extension → Remote Extension (Streamable HTTP) → paste the URL. Desktop: Extensions → Add custom extension. Or: goose session --with-streamable-http-extension <MCP URL>.

https://stabledomains.net/api/mcp
Goose docs →

n8n AI Agent

Workflow automation with AI agent nodes.

Add an MCP Client Tool node to your AI Agent, transport HTTP Streamable, endpoint:

https://stabledomains.net/api/mcp
n8n AI Agent docs →

Dify

Open-source LLM app and agent builder.

Integrations → Tools → MCP → add server (HTTP only) → paste the URL, name, and a stable Server ID.

https://stabledomains.net/api/mcp
Dify docs →

LibreChat

Self-hosted multi-model chat with MCP.

Add to librechat.yaml:

mcpServers:
  stabledomains:
    type: streamable-http
    url: https://stabledomains.net/api/mcp
LibreChat docs →

LangGraph / LangChain

Popular agent framework; MCP via langchain-mcp-adapters.

Python:

from langchain_mcp_adapters.client import MultiServerMCPClient
client = MultiServerMCPClient({"stabledomains": {"transport": "streamable_http", "url": "https://stabledomains.net/api/mcp"}})
tools = await client.get_tools()
LangGraph / LangChain docs →

CrewAI

Multi-agent crews in Python.

Python:

from crewai_tools import MCPServerAdapter
with MCPServerAdapter({"url": "https://stabledomains.net/api/mcp", "transport": "streamable-http"}) as tools:
    ...
CrewAI docs →

Vercel AI SDK

TypeScript agents on any model.

TypeScript:

const mcp = await createMCPClient({ transport: { type: "http", url: "https://stabledomains.net/api/mcp", headers: { Authorization: "Bearer sd_live_..." } } });
Vercel AI SDK docs →

MCP Registry

Clients that browse the official registry can find us by name.

Search for net.stabledomains/registrar.

net.stabledomains/registrar
MCP Registry docs →

Directories and read-only chat apps can use https://stabledomains.net/api/mcp/discover: search and pricing tools, with a checkout link instead of buy tools. Full tool list, prompts and safety rules: connect your agent · developers · skill.md.

StableDomains integrations | Install the domain MCP server in any agent