Spot domain market trends before they peak. This skill transforms raw keyword spike data into actionable investment signals by separating real multi-party demand from single-operator noise.
Raw keywords_trends(emerging) data typically returns 50-100 keywords — most are junk. This skill's value is extracting signal from noise:
This skill requires the DomainKits MCP connection and access to web_search.
keywords_trends, nrds, and other domain intelligence toolsNo additional API keys or environment variables are needed beyond the DomainKits connection.
Connect DomainKits via Settings → Connectors. The platform handles authentication automatically.
Add to your MCP config:
{
"mcpServers": {
"domainkits": {
"baseUrl": "https://api.domainkits.com/v1/mcp"
}
}
}
With API key (for higher limits):
{
"mcpServers": {
"domainkits": {
"baseUrl": "https://api.domainkits.com/v1/mcp",
"headers": {
"X-API-Key": "$env:DOMAINKITS_API_KEY"
}
}
}
}
Get your API key at https://domainkits.com
This skill orchestrates the following tools:
keywords_trends — Fetch emerging keyword spikes (DomainKits MCP)nrds — Search newly registered domains by keyword and position (DomainKits MCP)web_search — Investigate catalysts behind registration spikes (platform built-in)Optional follow-up tools (user-driven):
deleted — Recently dropped domainsexpired — Backorderable domainsaged — Domains listed for salekeyword_intel — Deep keyword analysisdomain_generator — Creative name ideasCall keywords_trends(type="emerging") to get keywords with registration volume spikes in the last 7-14 days. The tool returns per-keyword data including registration volume (w3/w4), com_ratio, forsale_pct, might_use_count, top_registrar, and other dimensions needed for Step 2 analysis.
The domain market is a multi-party ecosystem. After registering a domain, a person can only do one of three things with it:
forsale_pct.might_use_count — but this only indicates the domain was configured beyond default parking, not that it is genuinely in use. It could be a real project or a site farm. Only meaningful when registrar distribution is diverse.These three states account for every registered domain. Like any financial market, a healthy keyword market requires liquidity — active trading, not just ownership.
forsale_pct is the market's trading volume. If forsale is very low, market participation is low — this is not a healthy market signal. A keyword with high com_ratio, dispersed registrars, and high might_use_count but near-zero forsale has registrations but no market.
Two hard rules:
Two cross-cutting dimensions apply to all three participant types:
com_ratio — the "blue chip ratio." High com_ratio means participants are investing in .com — the most expensive TLD. Low com_ratio means activity is concentrated on cheap TLDs.top_registrar.pct — the "exchange concentration." Registrars are channels, not identity labels. High concentration (e.g., above 80%) on a single registrar reduces confidence that many independent parties are involved. A real multi-party market almost always shows distributed registrar usage.The core question for every keyword is: Is this data from "one person" or "a market"?
When analyzing a keyword, check whether each participant type is present. When a type is absent, ask why — the answer tells you what's really happening. When a type overwhelmingly dominates, ask whether that makes sense for a real market or whether it points to a single operator.
junk (single operator or missing participant roles) or healthy (genuine multi-party market)Summarize filtering results concisely:
List healthy keywords with key profile data:
llm — W3: 794 → W4: 979 (↑23%)
com_ratio: 82.6% | forsale: 36.8% | might_use: 63 | top_registrar: Unstoppable Domains 29.7%
Profile: Multi-party participation, .com dominant, mix of investment and usage. Healthy signal.
For each healthy keyword identified in Step 2, use web_search to investigate what is driving the registration spike. Domain registration spikes are predominantly driven by technology and internet industry events — product launches, AI model releases, platform announcements, viral open-source projects, regulatory changes, major acquisitions, or high-profile domain sales.
Profile analysis tells you WHETHER a signal is healthy. Catalyst research tells you WHY — and "why" determines whether the opportunity is worth pursuing.
For each healthy keyword, search with a technology lens. Prioritize the last 3 days — emerging spikes are almost always driven by very recent events. If nothing is found within 3 days, extend to 10 days maximum:
{keyword} news — look for product launches, funding rounds, open-source projects, platform announcements, regulatory changes{keyword} domain sold price — a high-value domain sale is the single strongest catalyst for registration spikesPresent catalyst findings as this table. The Source column must contain a real URL from web_search results. No URL = keyword does not appear in the table. Do not substitute with training data.
| Keyword | W4 | forsale% | Catalyst | Source |
|---------|-----|---------|----------|--------|
| molt | 576 | 17.2% | OpenClaw/Moltbot AI agent project | [CNBC](url) |
| nemo | 374 | 29.7% | Nvidia NeMo/NemoClaw, GTC 2026 | [TradingView](url) |
| llm | 768 | 28.6% | PrivateLLM.com sold $250K | [DomainInvesting](url) |
Rules:
This step bridges "macro trend" to "micro execution."
For each healthy keyword, call nrds to examine actual registration patterns:
nrds(keyword="<keyword>", position="start", tld="com", no_hyphen="true", sort="reg_date_desc", days_range="0-10")
nrds(keyword="<keyword>", position="end", tld="com", no_hyphen="true", sort="reg_date_desc", days_range="0-10")
position=start (e.g., llmtools.com, llmagent.ai) → keyword as category anchorposition=end (e.g., myllm.com, bestllm.io) → keyword as modifierperiod (registration term): 6+ years = serious project, 1 year = speculative trialprefix_tld_count: high = prefix registered across many TLDs = strong recognitionllm — NRDS Registration Analysis
position=start: 287 domains (llmtools, llmagent, llmchat, llmcode...)
position=end: 143 domains (myllm, bestllm, openllm, smartllm...)
Popular combos: agent (42), tools (28), chat (23), code (19), hub (15)
Market direction: Heavy llm+agent combinations suggest bullish sentiment on LLM Agent space
Quality: Short domains (<10 char) largely taken, 10-15 char range still has room
After presenting the trend analysis and NRDS findings, let the user know they can use DomainKits' other tools to explore domain opportunities for any keyword that interests them — such as deleted for recently dropped domains, expired for backorderable domains, aged for domains listed for sale, keyword_intel for deep keyword analysis, or domain_generator for creative name ideas. Let the user choose which keywords and directions to pursue.
Guest users can use this skill with limited daily search quota. Register a free account at https://domainkits.com to unlock higher search limits and access to all tools.
共 2 个版本