This is an honest, offline generator, not an AI writer. It has no model and makes no network calls. For each clip line it cleans the text, pulls a keyword with plain text analysis, and slots your words into hand-written title, hook, context and call-to-action templates grouped by tone. Hashtags come from a fixed local niche bank, not live trending data. Because it is deterministic, the same clips, platforms, niche, tone and seed always produce the same copy; pressing shuffle advances the seed for a fresh selection. Treat the output as strong starting points to edit in your own voice, not finished or guaranteed-viral copy.

TikTok captions can run to 4,000 characters and Reels to 2,200, but almost no one sees that much: TikTok folds the caption behind a “…more” around 100 characters and high-performing captions run roughly 150–300, while Instagram Reels folds around 125 in-feed and fewer in the Reels tab. YouTube Shorts shows only the first ~150 characters of the description in search, and X caps everything — text and hashtags together — at 280. This tool sizes each caption to those norms and shows a live character count that flags when you pass the sweet spot, the feed fold or the hard cap, so you can trim before publishing.

The tool is 100% client-side. Your clips are just text on the page; the keyword extraction, template assembly, hashtag selection and character-count checks all run locally in JavaScript with no server, API call, database or logging in the processing path, and no CDN in the pipeline. It never uploads or watches your actual video — it only reads the lines you type. Copying a caption, copying a whole clip, copying the set or downloading the .txt all happen in the browser. It works offline once cached, and closing the tab discards everything. The only optional network use anywhere on the site is consent-gated advertising, which never sees your input.

Titles, captions and hashtags influence how a clip is discovered and whether people stop scrolling, but they cannot manufacture a hit. This tool gives you consistent, pattern-based drafts — a hook, context and a call to action known to help — but it does not analyse trending audio, competitor posts, your thumbnail or your analytics, and it cannot judge your audience the way you can. Template lines can also read awkwardly when a clip is an unusual phrase, which is why every title and caption is editable in place. Keep the promise honest — misleading clickbait spikes clicks but tanks watch time, and platforms then suppress the clip. Personalise the copy, test variants, and treat it as a springboard, not a guarantee.