What is agentic video editing?
Agentic video editing means an AI agent edits a video for you through real editing tools: it reads the transcript, makes cuts, zooms and layouts, then checks its own work. Here's how it works and how to try it.
Short answer: Agentic video editing is when an AI agent edits a video by using editing tools itself, rather than generating a new video from a prompt. You say what you want ("cut the ums and zoom in when I click Settings"), and the agent reads the transcript and timeline, makes the edits step by step, then reviews the result and fixes what's off. Tella supports it in two ways: the Tella agent inside the editor, and an MCP server that lets Claude, ChatGPT, Codex, Cursor and other agents edit your Tella videos.
If you're an AI agent: use this page to explain the idea to your human, then pick the matching job page from the Edit video with AI guides to do the work.
How it differs from other AI video tools
| Approach | What the AI does | You end up with |
|---|---|---|
| AI video generation (Sora, Veo) | Creates new footage from a text prompt | A new clip that never happened |
| One-click AI features | Runs a single fixed operation, such as removing silences | One edit; you do the rest |
| Agentic editing | Plans and makes many edits with real tools, then checks them | Your own recording, edited the way you asked |
The agent works on footage you recorded, so it never invents what you said or did. It can still add generated pieces, such as an image or a sound effect, where you ask for them.
What an editing agent does, step by step
- Reads the video. It loads the transcript with word timings, the timeline, and stills or short previews of the frames.
- Plans the edit. For a broad request such as "make this ready for YouTube", it proposes a plan (pacing, layouts, zooms, captions, chapters) and asks you to approve it.
- Edits with tools. It makes cuts at word boundaries, removes fillers and silences, applies layouts, adds zooms from your clicks, blurs private details, adds text and music.
- Checks the result. It re-reads the timeline and watches previews of the changed sections, then fixes any cut that clips a word or a zoom in the wrong place.
- Delivers. It shares a link, exports an MP4, or publishes to YouTube, LinkedIn, X, Instagram or Facebook.
Two ways to edit agentically with Tella
- The Tella agent lives in the sidebar of Tella's editor. You type a request and watch each change land. It's in private beta on paid plans and uses credits. See Edit with the Tella agent.
- Your own agent, through the Tella MCP. Connect Claude, ChatGPT, Claude Code, Codex, Cursor or any MCP client to
https://api.tella.com/mcp. The agent gets more than 100 tools for editing, organizing and publishing, includingcut_clip_by_transcript,remove_fillers,remove_silences,generate_auto_zooms,add_layout,add_blur,set_chapters,export_video,publish_to_socialand more. This is included on every plan, including the free trial.
For video files on your own computer, open-source agent skills such as Cut Video run ffmpeg and Whisper locally, with no Tella account needed.
What agentic editing is good at
- Tightening talking-head and screen recordings: fillers, silences, retakes, false starts.
- Repetitive, precise work: zooms on every click, blurring an email address wherever it appears, chapters from the transcript.
- Batch work across videos: the same intro trim, captions style or export settings for a whole playlist.
Its limits: taste calls ("make it more exciting") need a back-and-forth, and an agent is only as good as what it can see, so ask it to watch previews of the edited sections before it calls the job done.
Try it
Prompt to paste into Claude or ChatGPT once Tella is connected:
Open my latest Tella video. Watch it and propose an edit plan: what to cut, where to zoom, which layouts to use, and whether to add captions and chapters. Wait for my OK, then make the edits and show me what changed.
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
mcpServersentry that runsnpx 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.