name-suggestions

name	name-suggestions
description	Generate domain and product name suggestions for a captured idea — themed across categories (descriptive, evocative, abstract, founder-named, compound) — and check basic domain/social-handle availability. Use when the user is ready to give an idea a working name.

Name & Domain Suggestions

Generate candidate names with light availability checks.

Steps

1. Read the idea and the latest formatted summary so the suggestions are grounded.
2. Generate name candidates across categories. Aim for ~20 total spread across:
   • Descriptive — what it does (Stripe → it routes payments; "RouteAPI")
   • Evocative — metaphor / vibe (Slack, Notion)
   • Abstract / coined — invented words (Twilio, Yelp)
   • Compound — two real words (Dropbox, GitHub)
   • Founder / proper noun — name-after style (Hewlett-Packard)
3. For each candidate, note:
   • Pronounceability and spelling-from-hearing test
   • Any obvious negative connotations or existing-brand collisions
   • Suggested TLD options (.com first preference, then .io / .ai / .so / niche TLDs)
4. Availability check — for the top 5–8 candidates:
   • Run whois <name>.com (and .io, .ai) to check basic registration status. Don't pretend this is definitive — ICANN data has lag and resellers obscure it.
   • Optionally use WebSearch to check if a current product / company already uses the name prominently.
   • For social handles, point the user at the relevant platforms (don't try to scrape availability — too brittle).
5. Present a ranked shortlist with the user's likely top 3.

Output

<workspace_path>/ideas/<slug>/exports/name-suggestions-<YYYYMMDD>.md containing the full slate, the shortlist, and availability notes.

Notes

• Domain availability is necessary but not sufficient. Cheap aftermarket domains are often worth the cost for a strong name.
• Trademark search is out of scope here — the user should consult a trademark attorney before committing. Note this in the output.
• Avoid names that are racist, sexist, or align poorly with the user's stated values from user-context.md.