Tutorial7 min read

How to Search for Domains With an AI Agent (Cross-Registrar Workflow)

An AI agent can brainstorm names, check availability, and compare prices across three registrars in a single conversation — if you install the right MCPs. This is the five-step workflow: pick registrars, install the MCPs, brainstorm candidates, fan out the availability checks, pick the cheapest source. Works best in Claude Desktop or Claude Code; ChatGPT supports a subset of the same flow.

Goal

Compare domain availability + price across registrars

Time

~15 min one-time setup, ~2 min per search

Prereq

Accounts at Cloudflare, Name.com, Porkbun

Steps

1

Pick the registrars the agent will search

Domain availability differs by registrar — not by which one "owns" the domain (a name is either taken globally or not), but by which TLDs each registrar carries and what they charge. The minimum useful set: Cloudflare (at-cost, common TLDs), Name.com (broad inventory, official MCP), Porkbun (at-cost, broad TLDs). Adding GoDaddy or Namecheap is only useful inside ChatGPT.

2

Install the three MCPs in your client

For Claude Desktop, add all three to claude_desktop_config.json. Each takes its own credential — Cloudflare uses an API token, Name.com uses a username + token pair, Porkbun uses an API key + secret. Keep all credentials in env blocks, never in args.

{
  "mcpServers": {
    "cloudflare": {
      "command": "npx",
      "args": ["-y", "@cloudflare/mcp-server-cloudflare"],
      "env": { "CLOUDFLARE_API_TOKEN": "your_token" }
    },
    "name-com": {
      "command": "npx",
      "args": ["-y", "namecom-mcp@latest"],
      "env": {
        "NAME_USERNAME": "your_user",
        "NAME_TOKEN": "your_token",
        "NAME_API_URL": "https://mcp.name.com"
      }
    },
    "porkbun": {
      "command": "uvx",
      "args": ["porkbun-mcp"],
      "env": {
        "PORKBUN_API_KEY": "pk1_...",
        "PORKBUN_SECRET_KEY": "sk1_..."
      }
    }
  }
}
3

Prompt the agent to brainstorm names first

Most domain searches start with a vague idea, not a known name. Ask the agent to brainstorm a list of candidates before any availability check — that turns a single registrar call into a structured workflow with five or ten parallel checks.

Brainstorm 10 domain name candidates for a new project
called "an AI agent that reviews pull requests". Mix
short, brandable names (.ai, .dev) with descriptive ones
(.so, .build). Return the list, then ask me to confirm
which ones to check.
4

Check availability across all three registrars

The agent should call each registrar MCP's availability tool for every candidate, then compose the results into a single table. The killer feature is price comparison — the same .com name typically costs the same at all three, but a .dev or .ai can vary by 30% or more.

For each candidate from step 3, use the Cloudflare, Name.com,
and Porkbun MCPs to check availability. Return a table with:
domain, available (yes/no), and price at each registrar
(USD, first year). Sort by price ascending.
5

Pick the winner and register it

Once the comparison table is in the conversation, pick the domain you want and the registrar to buy it from (usually the cheapest one that carries the TLD). Switch back to a single-registrar flow for the actual registration — the buy step is covered in the dedicated tutorial linked below.

I want to register codereview.dev from the cheapest source
in the table. Before calling register, restate the exact
domain, the cost, and which registrar you will use. Wait
for me to reply "yes" before proceeding.

Use read-only tokens for the search step

Availability checks are read-only operations at every registrar. Issue read-only API tokens for the search workflow and keep your write-scoped tokens locked in a password manager. When you decide to register a specific domain, swap to the write token, complete the purchase, swap back. The minor friction is worth the smaller blast radius.

Sample comparison table the agent should produce

This is what the output of step 4 should look like once the agent has fanned out across the three MCPs.

Agent output
Domain             Available   Cloudflare   Name.com   Porkbun
codereview.dev     yes         $12.18       $14.99     $11.98
codereview.ai      yes         —            $69.99     $54.99
codereview.io      yes         —            $39.99     $32.99
codereview.app     yes         $13.98       $14.99     $13.98
codereview.so      no          —            —          —

Cheapest available pick: codereview.dev at Porkbun ($11.98).

When it does not work

Agent only uses one MCP, ignores the others

Default behaviour when the prompt is ambiguous. Be explicit: "Use the Cloudflare, Name.com, and Porkbun MCPs in parallel" or list each tool by name in the request. The agent fans out reliably once it sees the explicit list.

Cloudflare returns "not supported" for a TLD

Cloudflare Registrar carries a curated TLD list — common .com / .org / .dev / .ai, but not many country TLDs and not every new gTLD. The "—" in the comparison table is the right behavior: the agent should still report availability for the other two registrars.

Name.com MCP hits the test environment

NAME_API_URL defaults to mcp.dev.name.com (safe). For real searches you need NAME_API_URL=https://mcp.name.com — the test env returns synthetic data that does not match reality. Verify the env var is set to production before trusting the prices.

Porkbun rate-limits the search

Porkbun's availability endpoint has a stricter rate limit than the others. For batches over 20 candidates, ask the agent to throttle — "check no more than 5 names at a time, pause briefly between batches."

FAQ

Why search across multiple registrars instead of one?

TLD pricing varies. A .ai domain can cost $50 at one registrar and $90 at another for the same first-year registration. Searching across three lets the agent surface the price spread before you commit — and for portfolios of 5+ domains a year, the difference compounds.

Can I do this in ChatGPT instead of Claude?

Partially. ChatGPT supports the Cloudflare MCP via Developer Mode connectors, and the GoDaddy and Namecheap ChatGPT Apps are first-party. Name.com and Porkbun do not have ChatGPT Apps — for those you would need the Developer Mode remote-MCP path with a hosted version of the server. Claude Desktop or Claude Code is the lower-friction setup for a multi-registrar workflow.

Does the agent need write access just to search?

No. All three MCPs separate "check availability" (read-only) from "register" (write). For pure search workflows, use read-only tokens — that caps the blast radius if the token leaks or the agent goes off the rails. Issue a write-scoped token only for the registration step itself.

How accurate are availability checks?

Availability is authoritative per-TLD via the registry, not cached. All three registrars hit the same upstream registries, so the answers agree. Pricing is a different story — each registrar sets its own retail price, so the agent has to query each one.

What if I want to add GoDaddy or Namecheap to the search?

You cannot from Claude or Cursor — GoDaddy and Namecheap publish their MCP integration only through the ChatGPT Apps directory. The workaround is to migrate the search workflow to ChatGPT (where all five work) or accept that the comparison is across Cloudflare / Name.com / Porkbun.

Does the agent get confused with three MCPs that all do the same thing?

Sometimes. The mitigation is in the system prompt: name each MCP explicitly when you want a specific one, and use "all three" or "each registrar" when you want the agent to fan out. Agents default to whichever tool was registered first if the prompt is ambiguous.

Once you picked a domain

The dedicated "buy a domain from Claude" tutorial walks through the registration step end-to-end, including the confirmation pattern that prevents accidental purchases.

More guides

Fundamentals

What Is MCP? A Plain-English Guide to Model Context Protocol

6 min read

Setup Guide

Best MCPs for Cursor in 2026 (Ranked + Setup)

8 min read

Setup Guide

Best MCPs for Claude Desktop in 2026 (Ranked + Setup)

9 min read

Setup Guide

Best MCPs for Claude Code in 2026 (Ranked + Setup)

8 min read

Setup Guide

Best MCPs for Windsurf in 2026 (Cascade-Ready Setup)

8 min read

Setup Guide

Best MCPs for VS Code in 2026 (Agent Mode + .vscode/mcp.json)

8 min read

Strategy

MCP Registry vs Curated Directory: Which Should You Use?

5 min read

Setup Guide

Best MCPs for ChatGPT: The Apps and Connectors Worth Installing

9 min read

Tutorial

How to Add an MCP Server to ChatGPT (Developer Mode + Apps Directory)

7 min read

Security

MCP Security: What to Know Before You Install

9 min read

Role Guide

Best MCPs for Marketers in 2026 (SEO, Email, Analytics)

8 min read

Strategy

Remote vs Local MCP Servers: When to Use Each

7 min read

Fundamentals

MCP vs Function Calling: What’s the Difference?

6 min read

Comparison

MCP Directories Compared: Top MCPs vs mcp.so vs PulseMCP vs mcp.directory

8 min read

Security

MCP Prompt Injection: How Tool-Calling Agents Get Hijacked

8 min read

Security

OAuth 2.1 for MCP: What the Spec Standardised and What You Need to Know

8 min read

Security

Sandboxing MCP Servers: Containers, Least Privilege, and Process Isolation

9 min read

Security

Rotating MCP Credentials: A Practical Guide for Leaks, Expiry, and Routine Hygiene

7 min read

Security

Least-Privilege Scoping for MCPs: How to Grant the Smallest Useful Permission

7 min read

Setup Guide

Best MCP Servers for Databases in 2026 (Ranked + Setup)

10 min read

Setup Guide

Best MCP Servers for Research in 2026 (Search, Scrape, Synthesize)

9 min read

Setup Guide

Best MCP Servers for Design-to-Code in 2026 (Figma → React)

9 min read

Setup Guide

Best MCP Servers for Domains in 2026 (Registrars + DNS)

9 min read

Tutorial

How to Buy a Domain From Claude (Cloudflare MCP, Step by Step)

6 min read