Domain Recommender
/install domain-recommender
Domain Recommender
Use this skill when the user gives:
- an AI application direction
- a keyword
- a short product sentence
and wants domain recommendations that are:
- SEO-friendly
- brandable
- realistically usable
- checked for current availability before you answer
Outcome
Return a short list of domain names that are both:
- semantically strong for the product
- verified as currently available or very likely available based on live checks
Do not stop at brainstorming names. The final answer must include an availability check.
Workflow
1. Understand the product and derive naming roots
Extract:
- the core problem
- the target audience
- the strongest search intent words
- the most natural noun/verb roots
Then expand into:
- singular/plural forms
- shortened roots
- phonetic or spelling variants when natural
- useful suffixes and prefixes
Example:
calculatorcan expand tocalc,calcu,calculator,calcuratorwritercan expand towrite,writer,scribe,draftmeeting notescan expand tonote,memo,meet,recap,brief
Read seo-brand-patterns.md for naming patterns and scoring heuristics.
2. Generate candidate names locally first
Use the bundled script to generate candidate domains quickly:
python3 scripts/generate_candidates.py "AI meeting notes app"
Optional:
python3 scripts/generate_candidates.py "AI tax calculator" --tlds com ai io app
The script is only a generator. It does not prove registration availability.
3. Rank candidates before checking availability
Prefer names that are:
- easy to spell
- short enough to remember
- semantically close to the user intent
- likely to help search relevance through the root word
- not awkward or obviously spammy
Avoid names that are:
- trademark-heavy or clearly derived from famous brands
- too long
- full of random hyphens or digits
- semantically vague
- awkward to pronounce
4. Verify availability with live checks
Availability is time-sensitive. You must verify it live before returning results.
Recommended order:
- Use live web checks on registrar or marketplace pages for the exact domain.
- Use DNS checks as a quick filter, not as the final proof.
- If you cannot confirm registrar availability, explicitly say it is only a likely-available fallback and explain why.
At minimum:
- check the exact domain live
- include the exact TLD
- prefer confirmed availability over speculation
Use dig only as a helper:
dig +short example.com
dig +short NS example.com
Important:
- no DNS record does not guarantee the domain is unregistered
- parked domains and for-sale domains may still resolve
- the final answer should prioritize registrar-confirmed availability
5. Return only the best shortlist
Default to 5 results unless the user asks for more.
For each result, include:
- domain
- why it fits the product
- naming logic
- current availability status
If nothing strong is available:
- say that directly
- offer the next-best alternatives
- widen the TLD set only after
.comand the most relevant TLDs are checked
Output Format
Use a concise structure:
- product interpretation
- shortlisted available domains
- optional backup domains
For each domain, prefer a compact line such as:
calcurator.ai— calculator root with a memorable spelling variant; confirmed available
If availability is not fully confirmed, label it clearly:
calcpilot.io— strong fit, DNS-clear, but registrar confirmation not obtained
TLD Preference
Default priority:
.com.ai.io.app.co
Adjust only when the product makes another TLD more natural.
Scope Boundaries
This skill is for domain recommendation and availability checking.
It is not for:
- trademark clearance
- legal opinion
- full brand identity work
You may mention obvious trademark risk heuristics, but do not present that as legal clearance.
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install domain-recommender - After installation, invoke the skill by name or use
/domain-recommender - Provide required inputs per the skill's parameter spec and get structured output
What is Domain Recommender?
Recommend SEO-friendly, brandable domain names for an AI product idea, then verify current availability before returning candidates. Use when the user provid... It is an AI Agent Skill for Claude Code / OpenClaw, with 103 downloads so far.
How do I install Domain Recommender?
Run "/install domain-recommender" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Domain Recommender free?
Yes, Domain Recommender is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Domain Recommender support?
Domain Recommender is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Domain Recommender?
It is built and maintained by daviddonecn (@daviddonecn); the current version is v0.1.0.