Three Video Ad Scripts from Your Product Brief

Three short video ad script drafts, written only from your facts

This tool writes scripts, not videos. You get three draft versions of a short ad with different hooks, each as a cut list with timed voice-over, on-screen text, a caption and SRT captions, ready for you to film, edit or hand to a creator.

The lines that get short ads rejected, or mislead the people who see them, sound normal in a draft: "the best bottle you'll ever own", "I tried it and it changed my life", "9 out of 10 users", "only today". Before you see the drafts, a script flags that kind of wording in every spoken line, on-screen text and caption, and the flagged lines are redrafted. You get a table of what changed and why.

The flag looks at wording only. It cannot tell whether a fact is true, so check every fact, number and name yourself before you publish.

What it looks like

A real run for a sample folding bottle: the first draft had a first-person review, a guarantee, an unproven "best", a made-up statistic, a price with no date, fake urgency, a before/after shot and a staged customer. Each was flagged and redrafted, and the final cut list is timed to 15 seconds.

What you send

• What the product or service is and does, the price and the date it applies, any offer with its real end date, who it is for, and the length (15, 30 or 60 seconds)
• (optional) facts with their source, real customer quotes you have permission to use, whether a creator posts it, platform, language

What you get (text and files)

• Three draft versions with different hooks (problem, demo, question, contrast, how-to, offer)
• A cut list per version: time, visual, voice-over timed to the cut, on-screen text, sound
• Captions that start with the ad label (#ad, Paid partnership, or your language's equivalent)
• claims_review.csv - every flagged phrase, the rule, the reason and the final wording
• ads.md, cuts.csv and SRT captions for each version

What the wording flag looks for

Guarantees and absolutes; "No.1", "best" and other rankings without a source; health, medical, weight or beauty effects; before/after; any first-person "I" or "my" voice that is not a real customer quote you supplied; numbers that are not in your facts; prices without a date; urgency without a real end date; earnings promises; staged "customer" scenes; a missing ad label. It covers English and Japanese wording.

What it does not do

• It never invents reviews, personal experiences, customer stories, results or specs. A fact you did not give stays a placeholder like [ADD: ...].
• The flag is a drafting aid, not legal or regulatory review, and does not verify facts or certify an ad as compliant. Regulated products are declined or kept to plain product facts.
• It makes no video, images or voice, and touches no ad account.