Weftly

Weftly Blog

From Raw Multicam Footage to a Published Video, Without Touching a Timeline

AI video editing with Claude and Weftly: how we took raw multicam footage to a published video with auto thumbnail, captions, chyrons, and filler removal.

Every creator knows the trap: you capture something meaningful, then lose a weekend to editing. This week we set out to prove a different workflow: one where the storyteller stays focused on the story and the tooling handles the rest.

The finished piece: thumbnail, captions, chyrons, end card, and the tightened cut, all produced by Weftly and Claude skills and plugins.

We started with our usual candid two-person conversation format about how The Woven Record began. The raw footage came from multiple camera angles, stitched into one longer recording by our multi-camera edit plugin: no manual syncing, no timeline juggling. That full recording was all we uploaded to Weftly.

From there, every step ran through Weftly and a handful of free Claude skills and plugins:

  • Weftly found and cut the clip. Its clip finder ranked the strongest moments in the full recording and surfaced this one, and the cutter pulled it out. The video you are watching is itself a Weftly clip, not a hand-trimmed export. Clips are one of our core products: find the moments, cut each for a few cents, no re-upload.
  • Weftly transcribed the clip into a word-level transcript, the backbone for everything downstream.
  • The thumbnail skill picked the most engaging frame and dropped it on a branded lower-third for YouTube.
  • A chyron surfaced our call to action, [email protected], exactly when it was spoken.
  • A “Made with Weftly” lower-third and a branded end card, built from our brand assets, closed the video.
  • Finally, the filler-remover skill used that transcript to cut the ums, uhs, and false starts, trimming seven seconds of dead weight.

The entire piece was produced by Weftly and our Claude skills and plugins: thumbnail, captions, overlays, end card, and the final cut. No editing suite. No render farm. Just a conversation in, and a publish-ready video out.

What’s shipped today, and what’s next. The free plugins linked above cover the Weftly essentials: transcription, summaries, clip finding and cutting, and skills for blog and social drafts. The video-craft steps you see in this post, multi-camera edit, thumbnail generation, chyrons, and editing video directly from the transcript, are on our roadmap. Here’s our promise: everything we used to make this video will be available to you too, either as a Weftly product or as a free Claude plugin. Nothing here stays an in-house-only trick. We post progress to the public plugins repo as features land, and you can sign up for updates at weftly.ai. Sign up in June and every job you run this month is 100% free.

This one’s personal. In the clip itself, we talk about why we started The Woven Record and why we had to scale, the same reason Weftly exists:

We built this because we needed to scale. We need to be able to go from raw video to published stories without losing a weekend or more.

If you’re a small creator, that’s the whole point: your job is to simplify the path from idea to audience. These tools exist for empowering creators to spend their hours where they matter: telling stories, not trimming clips.

Common questions

What tools made this video?
Weftly handled the clipping, transcription, and media processing, and Claude skills and plugins handled the rest: a multi-camera edit to build the source recording, Weftly's clip finder and cutter to pull this moment from it, a thumbnail generator, lower-third chyrons, a branded end card, and a filler-word remover. No traditional editing suite was used. Weftly also loudness-normalizes the audio on every clip it generates, to -14 LUFS / -1.5 dBTP, the level YouTube, Spotify, and podcast apps target. That means your clips play at a consistent, professional volume across platforms with no manual mixing, and the true-peak ceiling keeps loud moments from distorting.
Do I have to edit the video myself?
No. The point of this workflow is that you stay focused on the conversation. You provide the raw footage; Weftly and Claude produce the transcript, thumbnail, captions, overlays, end card, and the tightened final cut.
How does filler removal work without manual cutting?
Weftly returns a word-level transcript with per-word timestamps. The filler-remover skill finds the ums, uhs, and false starts, then the exact spans are cut from the video automatically. Here it trimmed about seven seconds, and the overlays were re-timed to match. Full editing by transcript, where you cut and rearrange the video just by editing the words, is next up on our roadmap, with a target release of mid-June 2026.

Try Weftly

Transcribe, summarize, and clip videos in minutes. Pay per job, no subscription.

Try it in the browser