FAQ¶

Q: Is there a free Opus Clip alternative without watermarks or credits?

Yes. video-edit-cli is open source and free, with no watermarks, per-clip credits, or clip expiry. Your AI coding agent runs it locally.

Q: How do I make YouTube Shorts from a long video automatically?

Install the video-edit-cli skill and ask your coding agent to make Shorts from the video. It transcribes, picks the strongest moments, reframes to 9:16, burns in captions, and renders each clip locally.

Q: How do I remove silences, filler words, and false starts from a podcast?

Ask your agent to edit the recording into a publishable episode. It cuts from a word-level transcript, so it distinguishes dramatic pauses from dead air and removes false starts without clipping the retake.

Q: Does my footage get uploaded anywhere?

No. Everything runs locally, including transcription. Source files are immutable and every derived file gets a provenance sidecar.

Can an AI coding agent like Claude Code really edit video?¶

Yes. video-edit-cli gives the agent fixed commands for transcription, cutting, reframing, subtitles, and audio mastering. A bundled skill teaches the agent to gather evidence, write an edit plan that explains each cut, then render and review. You ask in plain language; the agent edits on your machine.

Is there a free Opus Clip alternative without watermarks or credits?¶

video-edit-cli is open source and free, with no watermarks, per-clip credits, or clip expiry. Hosted clip apps charge $15–29 per month and set per-minute limits. An hour-long video can use much of a starter plan. Here, you pay only for your own compute.

How do I make YouTube Shorts from a long video automatically?¶

Install the skill (npx skills add computerlovetech/video-edit-cli --skill video-edit-cli) and ask your agent, for example: "Make clips for YouTube Shorts from this video, with burned-in subtitles." The agent transcribes the video, picks the strongest moments from the transcript, reframes to 9:16, burns in styled captions, and renders each clip. See example prompts.

How do I remove silences, filler words, and false starts from a podcast?¶

Ask your agent to "edit this recording into a publishable episode: cut the false starts, retakes, and long silences." Unlike threshold-based silence cutters, the agent works from a word-level transcript, so it can distinguish a dramatic pause from dead air and remove a false start without clipping the retake. The main edit workflow shows the full recipe.

Is this like Descript's text-based editing?¶

The editing model is similar: you cut video through its transcript. But your AI agent runs this free, local CLI instead of a paid desktop app. The agent reads the transcript and writes the cuts. A reviewable edit plan records each cut and its reason.

Does my footage get uploaded anywhere?¶

No. Everything — transcription included — runs locally. Source files are never modified, and every derived file gets a provenance sidecar recording exactly how it was made. Nothing leaves your machine unless you publish it.

What do I need installed?¶

Only ffmpeg and ffprobe. The skill has the agent install the CLI itself (uv tool install video-edit-cli) when it's missing, and video-edit-cli doctor verifies the rest per workflow.

Which coding agents does it work with?¶

Any agent that supports skills: Claude Code, Codex, Cursor, GitHub Copilot, Pi, Amp, and others. The CLI itself is plain shell + JSON, so any agent that can run commands can use it directly.

Can I trust the cuts the agent makes?¶

You can audit them. In each edit plan, every cut has a written reason. You or the agent reviews each cut before rendering. Provenance sidecars let you trace and reproduce any edit. There is no hidden "virality score."

Does it run on Windows?¶

The CLI supports macOS and Linux. Use WSL on Windows, which remains untested.