The order is handled
Reframe before captions, audio before silence-cutting — chain order changes the result, and the presets encode the right one. You say the outcome; the pipeline sequencing is not your problem.
You already have the footage — it's just landscape, quiet, and full of pauses. Claude runs it through ViralMint's editing chain: reframe to 9:16, clean the audio, cut the dead air, burn animated captions, watermark. One message, one output file.
Make this recording TikTok-ready: ~/Desktop/interview.mp4
…and Claude drives the whole pipeline below on your machine.
Reframe before captions, audio before silence-cutting — chain order changes the result, and the presets encode the right one. You say the outcome; the pipeline sequencing is not your problem.
A 2 GB screen recording never leaves your machine — FFmpeg, Whisper and the caption renderer run on your own hardware. No upload queue, no length cap, no watermark you didn't ask for.
The same skill covers 'make this postable' and 'just cut the silences'. Fourteen single-purpose editors are exposed to Claude individually, so small asks stay small (and cheap).
tool_chain runs several editors in the correct order in one call — tiktok_ready (reframe → audio → captions) for landscape talking-heads, studio_polish for already-vertical footage, podcast_to_short for long talking clips.
Face-tracked reframing to 9:16, loudness normalization and denoise, silence cutting and word-by-word animated captions all run on your machine with FFmpeg and Whisper.
Need just one thing? Claude calls the individual tool instead — auto b-roll, zoom pulses, sound effects, watermark, speed, a GIF export, SEO metadata or YouTube chapters.
wait_for_job returns the output path directly — the polished mp4 is on disk, ready to post, with no export step.
tool_chaintool_reframetool_audio_enhancetool_remove_silencetool_captionswait_for_job Download the free, open-source desktop app (macOS / Windows / Linux) and open it. It bundles the whole pipeline — yt-dlp, Whisper, FFmpeg — and the MCP server.
It shows your loopback endpoint and a bearer token, with a ready-to-paste Claude Code command and a JSON config snippet for Cursor / Claude Desktop.
Paste the command. Your client loads ViralMint's tools, the built-in workflow recipes (this skill is one) and read-only resources like your balance and recent videos.
Type the prompt above. Claude runs the recipe end-to-end against the local app and hands you the finished mp4 — no clicking through the UI.
| What matters | Typical paid agent tools | ViralMint |
|---|---|---|
| Agent tools / skills | Behind a paid tier | Free & open-source (AGPL) |
| Where it runs | Their cloud | Your machine (loopback MCP) |
| Your footage & data | Uploaded to their servers | Stays local — only AI calls leave |
| Pricing | Monthly subscription | Prepaid, pay-per-render at ~cost |
| Output watermark | On free tiers | Never |
| Script quality | Prompt → generic script | Grounded in real breakouts + your data |
highlighted column = clearer fit.
The tiktok_ready preset chains three steps: reframe 16:9 → 9:16 with face/motion tracking so the subject stays centered, audio enhancement (denoise + loudness normalization to broadcast level), then word-by-word animated captions in your chosen style.
Yes — that's the main use. Any local file path works: screen recordings, phone footage, Zoom exports, downloads. ViralMint-generated videos resolve via their library id.
Yes — 'viral' (yellow-highlight, chunked phrases), 'classic' (full sentences, bottom) or 'bold' (green-highlight, Impact), plus emoji styling. Claude passes caption_style through the chain.
Each editing step is a small flat prepaid fee (the processing itself runs on your machine — no cloud render minutes). Claude can quote the chain with tool_cost_estimate before running it.
Install the free ViralMint desktop app, connect it to Claude Code or Cursor, and run this skill — no subscription, no watermark, your footage never leaves your machine.