open source free · no paid AI needed · runs on your laptop →

an agentic video editor
that works without an agent.

Write a few lines of text and get a finished video: visuals, captions, music and your own cloned voice. Free, on your computer. Connect your AI agent when you want it to do even more.

$ pip install cardbox-video && cardbox
🎙️ not my real voice 🎬 made with cardbox ✍️ written as plain text ✂️ zero editing

This explainer was made with cardbox. The visuals are cardbox, the voice is my AI clone.

how it works

you write the words.
cardbox makes the video.

No paid AI needed. Write plain text, and cardbox draws every frame, speaks it in your voice, times the captions to every word, adds music and exports an MP4 that's ready to post. All free, on your own computer.

The cardbox app: a script on the left, format, theme and voice options on the right, and preview frames below
01

Open cardbox

Install once, then type cardbox. The app opens in your browser. Everything runs on your computer.

02

Write what you want to say

Plain text, one paragraph per beat. That's it: no prompts, no paid AI, no timeline.

03

Pick a look

Vertical, landscape or square. A theme, caption style and music. Drop in your own images or clips if you like.

04

Make the video

Preview every frame, then render. You get the MP4 in your own (cloned) voice, plus a caption ready to post.

two ways to use it

works without an agent.
does more with one.

the core · free

Without an agent

Write plain text, pick a look, make the video. No paid AI, no API keys, no account. It all runs on your computer.

  • Plain text in, finished MP4 out
  • Your cloned voice, or the computer's voice
  • Your own images and clips from a folder
  • Themes, formats, captions and music
optional · agentic

With your AI agent

Connect Claude, Cursor or any MCP app and let it do the heavy lifting, with even more control.

  • Writes the script from one sentence
  • Finds openly licensed images, with credits
  • Picks animated scenes and times them to words
  • Looks at preview frames and fixes them
made for

if you can write it,
you can post it.

🧠Explainer reelsFaceless tech, science and finance channels, one idea per video
🚀Product launchesTurn release notes into a 30-second changelog video
💼LinkedIn updatesWeekly wins and lessons, square and captioned for the feed
🎓Course lessonsMicro-lessons that sound like the teacher, without a studio
🛠️Dev tipsTerminal demos and how-tos with typing animations
📰Newsletters to videoPaste the issue, get a short for every story
what's inside

small tool.
serious output.

👂

It listens to itself

Cloned voices sometimes fumble a word. cardbox transcribes every line it speaks, compares it with your script, and re-records anything that comes out wrong. Automatically.

script: "Every line it speaks"
take 1: "A V line it speaks" → retry
take 2: "Every line it speaks" ✓
🎙️

Your voice, on your laptop

Read ten short lines, about a minute, and every video speaks in your voice. Cloned on your computer, with a consent line recorded first.

⏱️

Animations hit on the word

Write on: billion, and the counter starts the moment you say "billion".

🖼️

Images you can use

Openly licensed images only, from Openverse and Wikimedia Commons. Credits written for you.

🎵

Music, no copyright strikes

A soundtrack synthesized for every video, dipping under your voice.

🧩

13 animated scenes

Counters, charts, flows, terminals, AI chats, logo reveals, format morphs, photo push-ins and more, with 1,900 icons.

🎨

7 themes, 4 caption styles

Dark, neon, sunset, paper and more. Any accent color, sans or mono, captions that pop or stay minimal.

formats

one script.
every platform.

Change one word, format: landscape, and every beat re-lays itself out: headline beside the animation for YouTube, stacked for the LinkedIn feed.

A vertical frame
Vertical · Reels, Shorts, TikTok
A landscape frame
Landscape · YouTube, X, slides
A square frame on the light theme
Square · LinkedIn and Instagram feed
midnight ember forest neon sunset mono paper
the agentic part · optional

connect an agent.
it edits for you.

Add cardbox to Claude, Cursor or any MCP client. Your agent writes the script, looks at preview frames before rendering, finds images, helps you record your voice, and renders.

// claude_desktop_config.json · .cursor/mcp.json
{
  "mcpServers": {
    "cardbox": { "command": "cardbox", "args": ["mcp"] }
  }
}
make_videoplain text in, video out
preview_beata frame the agent can see
render_reelMP4 + caption + credits
search_imagesreusable licenses only
voice_recordopens the recording studio
create_reelfull control over scenes
for developers · go render api

run it as a service.

A Go server that queues renders, streams their progress live and hands back the video. One binary and SQLite.

  • Worker pool and bounded queue, with backpressure, cancellation and a graceful drain on shutdown
  • Live progress over Server-Sent Events, to any number of watchers
  • Idempotency keys; identical specs reuse the finished render
  • Signed webhooks (HMAC-SHA256) with retries and backoff
  • API keys, rate limits, request IDs, JSON logs, Prometheus metrics
# submit
$ curl -X POST localhost:8080/v1/renders \
    -H "Content-Type: application/yaml" \
    --data-binary @video.yaml
{"id": "rnd_e3016b6c59a9", "status": "queued"}

# watch it render, live
$ curl -N localhost:8080/v1/renders/rnd_e3016b6c59a9/events
event: progress  {"stage":"voice","done":0,"total":3}
event: progress  {"stage":"draw","done":2,"total":3}
event: status    {"status":"succeeded","seconds":11.4}

# download
$ curl -O localhost:8080/v1/renders/rnd_e3016b6c59a9/video

your turn.

Free and open source. No paid AI needed. Your voice never leaves your machine.