ReelPack Studio: Short Video, Brief to Publish-Ready Pack

Hand a video model and an editor a complete pack, and catch the risky parts before you publish.

ReelPack Studio takes one short video from brief to a pack you can hand to a video model and an editor, and checks it before you publish. It writes the hook options, script and shot list with you, then measures and checks the parts that can be measured.

What you get

• Hook lab: your hook variants scored by seven visible rules, with the passed and failed rules shown (a pattern count, not a retention prediction)
• Timed shot list: every line timed from your speaking pace, with total length and pacing checks
• Prompt pack: one prompt per shot that repeats the same character and style text word for word, so characters stay consistent, in a layout you choose
• Captions and cut list: an .srt file and a cut list with timecodes
• Pre-flight QA: absolute or unsupported claims, figures without a source, banned terms, real-person likeness without consent, music or assets without a license, AI-labelling, regulated topics without a disclaimer, platform length and text limits
• Metadata: title, description and hashtags checked against limits
• One export folder with every file, plus a JSON and Markdown report

Example (sample project)

A 7-shot project comes back as a 26.6 second pack with 42 spoken words and 11 caption cues. Pre-flight QA rates it high risk with 7 findings, for example a banned term in the description, a real person with no consent reference, music with no license, and an "80 percent" figure with no source. The hook lab ranks the variants by rules passed (6 of 7, 4 of 7, 3 of 7). All names and figures are fictional sample data.

Inputs

• Product facts with their sources, platform and target length, style, characters, assets with licenses, banned terms

How it stays accurate

A deterministic offline script does the timing, scoring, captions, prompts and checks, so the same project always gives the same pack. The agent never invents a fact, license, consent or source, and asks you when a finding needs your answer.

Limits

• It does not generate video or call any video service, and it cannot promise views
• Platform limits and prompt layouts are starting values you can edit. They are not verified against current platform or model rules
• The QA finds missing records and risky wording, not whether a claim is true or a license is valid
• It gives no legal, financial or medical advice

Data sharing

Only your project details and pages you link are used. No passwords, keys or accounts are needed. The script runs locally with no network calls. The text you give the agent is processed by the AI model provider running it: the model you choose in Download mode, or the model configured on this listing when run on Capafy. Keep your files on your own machine, and don't share anything you wouldn't give that provider.

Try it

"Build my attached project.json, then show the QA findings, the ranked hooks, the shot table and the total length."