Storm Research

The boring problem it kills

Big decisions made on one confident answer. Ask an AI "should I buy an AI receptionist?" and you get one angle — usually the vendor's, with statistics nobody checked. This skill asks the question five ways: it figures out which experts would actually disagree about your topic, has each one research it separately, maps where they contradict each other, then re-checks every number against the original source. The scary stat that traces back to a vendor's marketing page gets flagged, not repeated.

What the briefing gives you

See a real one

This is an unedited briefing the skill produced for the question every service business is hearing pitches about right now: "Should I let an AI answer my phone?" Note the verification banner up top — that tally is real.

🔒 What this touches: your question is researched on the public web by your AI assistant — no business data leaves your machine unless you put it in the question. Sources are cited with URLs and independently re-checked. Nothing is stored by this skill. The briefing is input for your decision, honestly labeled — including what it could not verify. The panel reads; you decide.

One honest note before you install

This is the heaviest skill in the library. It needs an AI assistant that can run research agents and save files — Claude Code or the Claude desktop app is the straight path (drop the folder in, say "storm research [topic]"). In a plain chat assistant it still works, just slower: the perspectives run one at a time. Either way, it's for decisions worth twenty minutes — not for looking up an address.