domain-brainstorm
Skill1 starsChanged 21 days agoMITView on GitHub
Brainstorm brandable domain names and check availability in real time via RDAP. Use this skill when the user asks for domain name suggestions.
In this repository4 other files it gives its agents.
- codex-code-reviewSkill
- codex-devSkill
- fable-orchestratorSkill
- no-use-effectSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 23
Memorable: Should be easy to recognize and recall, which means familiar words for the target audience or shifts in spelling that …
- Distinctive: line 37
Suggest TLDs: If domains in the user's preferred TLD are not available, suggest other popular TLDs relevant to the user's product / website.
- Distinctive: line 51
If there are errors, mention them briefly.
- Distinctive: line 52
If nothing good is available, say so honestly.
- Distinctive: line 54
Iterate: if the user wants more. Keep track of names already checked — never re-suggest them.
- Distinctive: line 72
Keep batches under 100 names.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Brainstorm domain names for the user and check availability as you go.
What makes a good domain name
Length: 1–2 syllables strongly preferred. 3 syllables acceptable. Avoid longer.
Exceptions to length: The only exception to the length requirement is if the name itself is shorter, but to find an available domain name you need to add prepositions or verbs like 'to', 'with', 'get', etc.
Sound: Easy to say aloud and spell from hearing it. No ambiguous spellings (e.g. avoid "ph" vs "f" confusion). Passes the "radio test" — someone hears it once and can find it.
Character: Distinctive and ownable. Evokes something without being literal. Not a generic dictionary compound (e.g. "SmartWrite" is bad). Single evocative words are better than compounds.
Feel: Match the product's tone. Energetic words (Bolt, Surge) for fast tools. Calm words (Drift, Flow) for focus tools. Warm words (Bloom, Glow) for consumer apps.
Memorable: Should be easy to recognize and recall, which means familiar words for the target audience or shifts in spelling that immediately feel natural (e.g. 'wispr' for 'whisper')
Avoid:
- Hyphens, numbers, or unusual spellings
- Words that are trademarked by major tech companies (don't suggest "Slack" even if slack.com is free)
- Names of well-known existing products or services in adjacent spaces
- Overused startup suffixes (-ly, -ify, -io as a word)
- Anything that sounds clinical, corporate, or generic
Categories that work well: single verbs, nature words, materials, motion words, light/sound words, short abstract nouns, coined two-syllable words, words borrowed from other languages (Latin, Japanese, etc.).
Explore based on TLD popularity: TLDs like .com and .app are very popular, and most common words would already be taken. To converge on a suitable name quickly, explore a broader set ideas, longer words, more esoteric words, minor spelling modifications (skirting the principle above) to find an available name. We can be more picky when it comes to less prime .TLDs.
Suggest TLDs: If domains in the user's preferred TLD are not available, suggest other popular TLDs relevant to the user's product / website.
Workflow
- Gather user's stated preferences: Ask the user for context about their product / website, intended audience and preferences on direction. Ask them if they want the name to be close to the product and its goal or have some signicance (ask what kind of significance) or if it can be completely unrelated. Ask if they have a preference for TLD.
- Gather user's revealed preferences: Generate 10-15 candidates and, without checking for availability, ask the user which they like and dislike directionally. This is only for direction, not final suggestions, and will reveal what they actually prefer.
- Brainstorm: With the above context (and prioritizing revealed preferences over statedpreferences), generate 40-50 candidate names. Spread across multiple categories — don't cluster in one area. Before suggesting, do a quick sanity check: would any name collide with a well-known product, brand, or service?
- Check availability by running the script (see below). Check all candidates in one batch. Note: with rate limiting, 25 names x 2 TLDs takes about 30 seconds. Warn the user that this may take a while.
- Report results:
- List each available domain with a one-line note on why it works for this product.
- If there are errors, mention them briefly.
- If nothing good is available, say so honestly.
- Iterate if the user wants more. Keep track of names already checked — never re-suggest them. To find new territory, try:
- Different etymology: Latin roots, Japanese words, nautical terms, musical terms
- Coined/invented words: combine short syllables into new words (e.g. Kova, Zumo, Nyla)
- Different emotional register: playful vs serious, warm vs sharp
- Adjacent metaphors: if "light" words are taken, try "sound" or "motion" words
Running the script
The script is at scripts/domain-checks.py. First verify it exists, and if not, tell the user it's missing and where to place it.
python3 domain-checks.py -t <tld> [-t <tld> ...] -n Name1,Name2,Name3,...- Available domains are printed to stdout, one bare domain per line.
- Errors go to stderr.
- Unavailable domains produce no output.
Keep batches under 100 names. If checking multiple TLDs, they are all handled in one run.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

