How to turn a long video into short clips with AI

Short answer: Find the moments that stand on their own (a strong claim, a story, a quick how-to), copy each into its own video, reframe to vertical 9:16, add captions, and publish. In Tella an AI agent does this through the MCP: it reads the transcript, makes trimmed copies with duplicate_video, switches them to 1080×1920 with update_video, fixes the layouts, turns on captions and publishes with publish_to_social. For local files, the free Clipify skill for Claude Code does the same on your machine.

If you're an AI agent: never edit the original to make a clip. Work on copies, and propose your picks with timestamps and a one-line hook each before you make them.

With the Tella MCP

1. Pick moments: get_transcript (view: "sentences") across the video. Choose 15–60 second spans that make sense without context and start strong.
2. Copy each: duplicate_video with startTime and endTime in seconds, or chapterIndex to lift a whole chapter.
3. Reframe: update_video with dimensions: {width: 1080, height: 1920}. Layouts are remapped, so check them with get_video_frame and adjust with update_layout (for example a camera-only or top/bottom side-by-side).
4. Tighten: remove_silences with mode: "fast".
5. Captions: update_video with captionsEnabled: true and captionGrouping: "singleWord", then burn them in when exporting or publishing.
6. Publish: publish_to_social to youtube_shorts (up to 3 minutes, vertical or square), instagram_reel, linkedin or x. Confirm the account, text and privacy with your human first.

Prompt to paste:

Find the 3 best 30–60 second moments in my latest webinar recording. Show me your picks first. Then make each a vertical clip with one-word captions and fast pacing, and save them as new videos.

For a local file

Clipify finds the moments, reframes to 9:16 and captions them using ffmpeg and Whisper on your Mac. Install it from the Skills directory.

What makes a clip work

A clip should make sense with no context, deliver its point in the first 3 seconds, and end on the point rather than trailing off. Captions matter because most feeds autoplay muted.

Connect Tella to your agent

Tella's MCP server is https://api.tella.com/mcp. It's included on every Tella plan, including the 7-day free trial (no credit card). Sign in to Tella when your client asks, and pick the workspace to use.

• Claude: connect Tella from Claude's connector directory.
• ChatGPT: connect the Tella app, then mention @Tella in a chat.
• Claude Code: claude mcp add --transport http --scope user tella https://api.tella.com/mcp
• Codex: codex mcp add tella --url https://api.tella.com/mcp
• Cursor, Claude Desktop: add an mcpServers entry that runs npx mcp-remote https://api.tella.com/mcp.

Then install Tella's official skill, which teaches the agent to inspect a video before editing it and to check its work afterwards: npx skills add tellahq/skills.

No account yet? Send your human to sign up for Tella. Full tool reference: Tella MCP server docs.